In the ever-evolving landscape of machine learning and computer graphics, the introduction of Geometry-Informed Neural Networks (GINNs) marks a significant milestone. Developed by Arturs Berzins, Andreas Radler, Sebastian Sanokowski, Sepp Hochreiter, and Johannes Brandstetter, GINNs offer a novel approach to training shape generative models without relying on extensive datasets. This article delves into the core concepts and implications of GINNs, shedding light on their potential to transform various domains where data scarcity has been a persistent challenge.
The Challenge of Data Scarcity
The traditional approach to training neural networks, particularly in the realm of shape generation, heavily relies on large, annotated datasets. These datasets provide the necessary examples for the network to learn and generalize patterns. However, in fields like computer graphics, design, and engineering, acquiring such extensive datasets is often impractical. The lack of available data hampers the application of state-of-the-art supervised learning methods, necessitating alternative strategies.
Introducing Geometry-Informed Neural Networks
Geometry-Informed Neural Networks (GINNs) present a paradigm shift by enabling the training of shape generative models without any data. The core idea behind GINNs involves three key components:
Learning Under Constraints: GINNs leverage geometric constraints inherent to the shapes being modeled. These constraints guide the learning process, ensuring that the generated shapes adhere to the desired geometric properties.
Neural Fields as a Representation: Instead of relying on discrete data points, GINNs utilize neural fields. Neural fields offer a continuous representation of shapes, making them well-suited for capturing intricate geometric details.
Generating Diverse Solutions: One of the standout features of GINNs is their ability to generate multiple solutions for under-determined problems. This capability is crucial in scenarios where a single correct solution does not exist, allowing for a broader exploration of the solution space.
Credit: Tesfu Assefa
Applications and Results
The researchers applied GINNs to a variety of two and three-dimensional problems, each with increasing levels of complexity. The results were promising, demonstrating the feasibility of training shape generative models in a data-free setting. This breakthrough has significant implications for several fields:
Computer Graphics: Artists and designers can leverage GINNs to create complex shapes and models without needing extensive datasets. This could streamline the creative process and reduce the dependency on pre-existing data.
Engineering: Engineers can utilize GINNs to design and optimize structures where obtaining a comprehensive dataset is challenging. The ability to generate diverse solutions allows for innovative approaches to problem-solving.
Medical Imaging: In medical fields where annotated datasets are scarce, GINNs can assist in generating accurate models of anatomical structures, aiding in diagnosis and treatment planning.
Future Directions
The introduction of GINNs opens several exciting research directions. The potential to expand the application of generative models into domains with sparse data is particularly noteworthy. Future research could focus on refining the techniques used in GINNs, exploring new applications, and integrating GINNs with other machine learning paradigms to further enhance their capabilities.
Conclusion
Geometry-Informed Neural Networks represent a groundbreaking advancement in the field of shape generation. By enabling the training of generative models without relying on extensive datasets, GINNs address a critical limitation in current machine learning methodologies. The work of Berzins, Radler, Sanokowski, Hochreiter, and Brandstetter paves the way for innovative applications across various domains, highlighting the transformative potential of this new paradigm.
For those interested in exploring the detailed mechanics and applications of GINNs, the original research paper is available here. This pioneering work is poised to inspire further research and development in the exciting intersection of geometry and neural networks.
Reference
Berzins, A., Radler, A., Sanokowski, S., Hochreiter, S., & Brandstetter, J. (2024, February 21). Geometry-Informed neural networks. arXiv.org. https://arxiv.org/abs/2402.14009
Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter.
In a video released in April 2022, German startup Volocopter stated their goal was to get their flying car zipping around the skies over the 2024 Summer Olympics.
Volocopter is one of hundreds of start-ups pursuing advanced air mobility – essentially flying cars – and their vehicle is called the Volocity. Publicly announcing timelines in technology is usually considered a bad idea, and as late as February 2024, Politico rubbished their bold claim.
Volocopter can now gloat at the doubters. Last week they announced Olympic success. The technology and legal permissions will be ready for the Paris Olympics, the opening ceremony of which is today, 26 July. People will indeed be able to use flying taxis to hop around Paris during the upcoming Olympics, and President Macron is invited to go for a spin.
It is an interesting moment for electric aviation. Battery technology is boldly tearing forward with sodium-ion, solid-state, silicon anode, and other improvements. Cars, boats, and trains are switching to electric motors. “The electrification of aircraft is leading to many opportunities to develop fundamentally new configurations and to take advantage of distributed and mechanically disconnected propulsion”, one paper says. At the same time: drones have established themselves as a major industry, thanks largely to improvements in computer technology that can control many rotors at once. Most consumer and military drones are electric but unmanned, but it’s time to revisit the question of 1950s futurists: When can I get my flying car?
Credit: Conor O’Higgins using Stable Diffusion
eVTOL and eSTOL
There are two approaches to building flying cars: eVTOL and eSTOL. These stand for ‘electric vertical takeoff and landing’ and ‘electric short takeoff and landing’.
Look at a map of a city and you’ll see that in most cities, airports are the largest plots of land dedicated to any single purpose. There are two reasons for this: space and noise. Planes need long takeoff and landing distances (1.5km or 2km), and they are noisy, so they have to be set apart from homes and schools and stuff. These limitations often force airports outside of cities, and that adds an hour to your trip to Spain.
Dublin airport (outlined in orange) compared to Dublin. Note that the airport is huge, maybe 5% the size of the city herself, and is set apart from the city. (Credit: OpenStreetMap)
eVTOLs promise a quieter flying machine that can take off and land vertically (say on a roof). eSTOLs also aiming to run with quiet electric motors, and to take off and land a very short runway. Whereas commercial planes need 1.5-2km runways, eSTOLs aim to need “between 45 and 90 meters of runway” according to the COO of Electra.aero.
This video from Electra beautifully illustrates what an eSTOL should be: it’s simply a plane, but a battery-powered one optimised for urban deployment:
Flying with rotors, flying with wings, a bit of both
Why would anyone build eSTOLs for short runways when eVTOLs with zero runway are an option? It’s because the engineering is significantly simpler: an eSTOL uses familiar airplane technology: it drives forward to create lift under its wings and take to the sky. It is an adapted plane. This allows engineers to use well-understood principles and try to win the race-to-market.
eVTOLs are a little more complex. Most of them (Volocopter’s Volocity is one of the exceptions) require engineers to think about two phases: one where rotors or ducts pick it up and set it down for the vertical takeoff/landing, and another fixed-wing mode for cruising around. This second plane-like flight is quieter and more energy-efficient than the drone-like one.
Little planes or big drones?
Therefore eSTOLs should theoretically be simpler and cheaper than eVTOLs, but in this bubbling, surging industry, design philosophies abound. The Volocity by Volocopter that will share the skies with the Olympic polevaulters is an eVTOL with a single propulsion-type. It takes off and lands vertically like a drone, and then – unlike its competitors Joby, Beta Technologies, Lilium etc. – it continues to fly around in the manner of a drone. The simplicity of this design philosophy got the Volocopter Volocity to market before their competitors. Whereas eSTOLs are small planes, Volocopter’s machines are big drones.
A good Forbes article said, “It’s been estimated that around 300 different companies are trying to build new “flying car” electric VTOL aircraft for the anticipated revolution, and there are almost as many different design philosophies. Most are opting for hybrid designs that feature rotors for vertical takeoff and landing, but regular fixed wings for horizontal flight. There’s a good reason for that — fixed wing flight is much more efficient, and for electric aircraft, battery weight is the key issue, and that makes efficiency really important. In spite of this, one of the companies furthest along in actual deployment is using a much more basic electric multirotor design with no fixed wings. It’s effectively a human sized drone”.
So Volocopter’s approach appears to have won on simplicity and quick deployment. The tradeoffs? Range and noise. Flying with rotors is less energy-efficient than fixed-wing flight, so you drain your battery quicker. The Volocity has a range of 35km.
Is Volocity’s 35km range a problem? Not really. You only need your vehicle’s range to be as long as your trip. Hundreds of kilometers is for city-to-city. A range of 35km is grand for hopping around one city.
For comparison, a video released by Joby, another of a many eVTOL startups, in 2021, says they flew 154.6 miles (248.8km) in a test. That was their battery-powered version; for longer ranges, they feel hydrogen is the way to go, and last month (June 2024) flew their hydrogen eVTOL for 523 miles.
An invaluable resource for tracking and comparing these projects is the Advanced Air Mobility Reality Index. Tech news is so full of hype and self-awarded “breakthroughs” that it’s handy to have someone independent keeping track of the projects and their technology readiness levels. It provides a league table (Volocopter is currently winning) that usefully lists the intended use-cases (flying taxi being the most common) and whether the bird is piloted or autonomous.
The Volocity aircraft, made by Volocopter, flying at the Paris Air Show in 2023. (Credit: Photo by user Ibex73 from Wikimedia Commons)
Noise
I used to write about breaking tech that would change the world. Now I write about breaking tech that won’t make such a racket. I must be getting old.
The noise made by helicopters is a major nuisance and has limited them to very infrequent flying. And if you’ve ever been near a drone, you know that their noise is not a detail.
Noise is the hardest engineering problem in eVTOL/eSTOL, and, in my opinion, the biggest shortcoming of Volocopter’s Olympic success. Electric cars run much quieter than combustion ones, but no such luck with aircraft. The noise from helicopters, drones, and airplanes doesn’t mostly come from the engine, but mostly from the interaction of the craft and the air (Likewise, electric cars are actually just as noisy as combustion cars at high speed, because wheel-noise is more important than engine-noise).
A valuable paper on low-noise electric aircraft puts it this way: “Although these aircraft use quiet electric motors instead of noisier combustion engines, this is not likely to have a significant effect on the overall noise radiation of the vehicle, because the noise of rotor and propeller driven aircraft is generally dominated by the aerodynamically-generated noise of the rotating blades. Instead, the main acoustic impacts of electrification are a result of the new freedoms of electric propulsion, especially distributed electric propulsion, offered to the aircraft designer”.
The fluid dynamics of why flying machines are noisy is extremely complex. eVTOLs spin their rotors at slower speeds than helicopters do to avoid the loud thwap-thwap noise that helicopters make. Yet turbulent flows bumping the body of the bird cannot be fully avoided, and complex multirotor designs send vortexes all over the place, including into collusion with their neighbouring rotors, creating noise.
Progress is being made on these problems (for example, a March 2023 paper found that using six blades instead of four lowered noise by 5 to 8 decibels, losing only a 3.5% thrust in the process. Volocopter say they are using “the lowest disc loading currently on the market… and a low RPM (revolutions per minute) rate” to reduce noise (‘Disc loading’ is the ratio of the bird’s weight to the area of its rotors; Volocopter positively bristles with rotors.)
This detailed fluid dynamics work can chip away at the noise problem – sensible work for centibel gains – but the reality of the engineering says don’t expect a masterstroke that will suddenly make eVTOLs 100× quieter. The only technology that may be an exception is the Lilium Jet. Instead of rotors, it uses 36 small ducted fans (so that the ducts trap sound), and is an undeniably beautiful-looking product:
Lilium sits at a mediocre 11th on the ‘Reality Index’, and until independent tests have certified the noise-levels of each experimental eVTOL, we have to be cautious.
A loud note of caution
I agree with the need for electric urban mobility. But let me question the need for flight.
‘Energy-efficient flight’ is, to some extent, a contradiction in terms. Under 0.25% of international freight is transported by plane (most goes on ships), and there’s a reason for that: energy costs. When you have to expend energy to do move a load from A to B, why also expend energy fighting gravity? You need a justification. Traffic might be a justification, but there are simpler ways to solve that (such as bicycle lanes). Emergency vehicles such as ambulances should fly: they have an excellent justification.
The noise issue is the hardest engineering problem around flying cars. Safety can be solved. Autonomy can be solved. eVTOL companies are quick to make impressive claims about their quiet birds, so we need an independent agency inspecting them for sound and publishing data. As this data is missing, and as I haven’t had the chance to get up close, we have to use some guesswork to figure out how loud they are.
I lean towards skepticism, and suspect that Volocopter have not made much progress on the sound problem. The approval from the City of Paris means they have achieved good safety standards, but it comes with limitations: they aren’t just allowed zip anywhere at any time. The authorities want them flying a limited number of flights at constrained times of the day. Volocopter’s website says they will “map out routes inside the city that ensure the Volocopter aircraft do not generate a cacophony that exceeds the city’s permitted noise levels. Part of its approach will involve flying at specific times of day.” This strongly implies there is quite a lot of noise. The most skeptical interpretation is to say they’ve simply built a small helicopter: those are also approved to fly low flight volumes at major events like the Olympics.
Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter.
Blockchain technology, initially conceptualized for digital currencies, has rapidly expanded into various sectors, including agriculture and food supply chains. This innovation offers a new level of transparency and efficiency, addressing long-standing issues in these sectors. This article explores the transformative impact of blockchain on agriculture and food supply chains, highlighting ongoing projects, challenges, and future potential.
The Need for Blockchain in Agriculture and Food Supply Chains
The traditional agriculture and food supply chains are often plagued by inefficiencies, lack of transparency, and trust issues among stakeholders. Complex and paper-heavy processes, risks of fraud, and high operational costs are common challenges. Blockchain technology offers a solution by providing a decentralized, immutable ledger that enhances transparency, traceability, and trust across the supply chain.
Blockchain Applications in Agriculture
Blockchain technology is being used to track the journey of food products from farm to table, ensuring each step is transparent and verifiable. Here are some key applications:
Food Traceability: Blockchain enables detailed tracking of food products through the supply chain. For instance, Walmart and IBM have used blockchain to trace the origin of mangoes, reducing the time needed to track the fruit from six days to a few seconds. This capability is crucial for food safety, enabling quick identification and removal of contaminated products from the market.
Supporting Small Farmers: Blockchain can help small farmers by providing them with better access to markets and financial services. Platforms like AgriLedger and FarmShare use blockchain to increase trust among small farmer cooperatives, facilitate fair pricing, and improve market access.
Food Safety and Quality Assurance: Blockchain, combined with IoT devices, can monitor and record conditions throughout the supply chain, ensuring products are stored and transported under optimal conditions. This integration helps in maintaining food quality and safety, preventing losses due to spoilage.
Reducing Food Waste: Blockchain can optimize supply chain operations, reducing waste and improving efficiency. By providing real-time data on inventory and demand, blockchain helps in better planning and resource allocation.
Challenges and Barriers
Despite its potential, blockchain adoption in agriculture and food supply chains faces several challenges:
Technical Barriers: Implementing blockchain requires significant technical infrastructure and expertise, which can be a barrier for small and medium-sized enterprises (SMEs) and farmers in developing countries.
Regulatory and Policy Issues: The lack of standardized regulations and policies for blockchain technology can hinder its adoption. Governments and regulatory bodies need to create a supportive framework to facilitate blockchain integration.
Scalability and Interoperability: Blockchain systems must handle large volumes of transactions efficiently. Scalability and interoperability with existing systems are critical for widespread adoption.
Cost and Accessibility: The cost of implementing blockchain technology and the required digital literacy can be prohibitive for some stakeholders, particularly in developing regions.
Case Studies and Success Stories
Several successful blockchain projects in agriculture and food supply chains demonstrate the technology’s potential:
AgriDigital: This platform executed the world’s first sale of 23.46 tons of grain on a blockchain in 2016. Since then, it has transacted over 1.6 million tons of grain, involving $360 million in grower payments. AgriDigital aims to build trusted and efficient agricultural supply chains using blockchain.
Louis Dreyfus Company (LDC): LDC conducted the first blockchain-based agricultural commodity trade, involving a shipment of soybeans from the US to China. The use of blockchain reduced document processing time to a fifth of the usual time, demonstrating its efficiency.
Carrefour: The European grocer uses blockchain to verify standards and trace food origins for various products, including meat, fish, fruits, vegetables, and dairy. This initiative ensures transparency and boosts consumer trust.
Credit: Tesfu Assefa
Future Prospects
The future of blockchain in agriculture and food supply chains looks promising. As technology matures, it will likely overcome current challenges, leading to broader adoption. Key areas for future development include:
Enhanced Integration with IoT: Combining blockchain with IoT devices can provide real-time monitoring and data collection, further improving supply chain transparency and efficiency.
Smart Contracts: The use of smart contracts can automate transactions and enforce agreements, reducing the need for intermediaries and enhancing trust among stakeholders.
Global Standards and Regulations: Establishing global standards and regulations will be crucial for blockchain’s widespread adoption in agriculture and food supply chains.
Education and Training: Increasing awareness and providing training on blockchain technology will help farmers and SMEs leverage its benefits effectively.
Conclusion
Blockchain technology holds significant potential to revolutionize agriculture and food supply chains by enhancing transparency, traceability, and efficiency. While challenges remain, ongoing projects and future developments indicate a bright future for blockchain in these sectors. By addressing current barriers and fostering innovation, blockchain can create more sustainable, trustworthy, and efficient food supply chains.
The cryptocurrency industry faced significant security challenges in Q2 2024 – and it failed some. Let’s look at the latest reports from the two leading crypto security firms: Immunefi and Hacken. The analyses paint a concerning picture of the current landscape, highlighting both familiar vulnerabilities and emerging trends. Data reveals a substantial increase in successful attacks which raises alarm bells about the need for improved security measures across the crypto ecosystem.
According to Immunefi, Q2 2024 saw a staggering $572.7 million lost to hacks and frauds across 72 incidents, representing a dramatic 112% increase compared to Q2 2023. Hacks continued to be the predominant cause of losses in the crypto space, with the vast majority of funds stolen through direct exploits rather than frauds or scams. This can be attributed to less awareness when markets trend upwards, which make it easier for bad actors to exploit newer users and stretched protocols.
Major Incidents
Two major incidents stood out in Q2, accounting for over 60% of total losses. The largest hack targeted DMM Bitcoin, a Japanese crypto exchange, resulting in a massive $305 million theft. This was followed by an attack on BtcTurk, Turkey’s largest cryptocurrency exchange, which suffered a $55 million loss in a cyberattack. These high-profile incidents highlight the potential vulnerabilities in even well-established exchanges and the devastating impact of successful attacks.
Shift in Attack Focus: CeFi vs. DeFi
Q2 2024 saw a significant shift in attacker focus, with Centralized Finance (CeFi) platforms bearing the brunt of attacks. CeFi losses totaled $401.4 million, accounting for 71% of all funds lost. This marks a massive 984% increase compared to Q2 2023. In contrast, Decentralized Finance (DeFi) platforms saw a 25% decrease in losses compared to the same period last year. This shift suggests that attackers may be finding centralized platforms to be more lucrative targets, possibly due to larger pools of concentrated funds.
Most Targeted Chains
Ethereum and BNB Chain remained the primary targets for attackers, with Ethereum suffering 34 incidents and BNB Chain experiencing 18.
Ethereum’s dominance as the most targeted chain highlights the ongoing need for heightened security measures in its ecosystem, especially as its total value locked (TVL) has grown significantly over the past year.
CeFi Accountable for the Biggest Losses (Credit: Hacken)
The types of attacks employed by malicious actors varied, but access control issues caused the highest losses at $397.2 million. Price oracle issues and flash loan attacks also contributed significantly to the overall losses. This breakdown helps identify areas where security measures need to be strengthened across the industry, providing valuable insights for both developers and security professionals.
Comparison to Previous Periods
The big increase in losses from Q2 2023 to Q2 2024 is worrying, especially considering the growth in total value locked across the crypto ecosystem. While the overall DeFi TVL tripled from about $50 billion to $150 billion by June 1, losses grew even faster.
It’s worth noting that despite fewer individual hacks compared to Q1 2024, the severity and financial impact of Q2’s attacks were significantly higher, indicating a trend towards more sophisticated and damaging exploits.
Implications for the Industry
The major hacks targeting CeFi platforms highlight the need for enhanced security measures in centralized systems. As the crypto ecosystem grows, maintaining security becomes increasingly challenging, and it will get worse if and when the 2024/2025 bull market returns. Projects must balance the desire for rapid growth with the need for robust security measures.
The industry may need to develop more comprehensive insurance solutions and standardized recovery protocols to soften the blows dealt by large-scale hacks. Additionally, these high-profile incidents may lead to increased regulatory scrutiny, potentially resulting in stricter oversight of crypto platforms, especially centralized exchanges.
Security Measures and Best Practices
Given the persistent threat of hacks and exploits, individual users and investors should take proactive steps to secure their assets. Some essential measures include:
Using hardware wallets for long-term storage
diversifying holdings across multiple platforms,
enabling two-factor authentication
staying informed about the latest security best practices
By adopting these proactive steps, users can significantly reduce their risk exposure in the face of evolving security threats.
Positive Developments
Despite the concerning trends, there’s some good news in crypto security. The industry is showing an improved ability to recover stolen funds, with about 5% of the total losses in Q2 2024 being recovered.
This represents a slight improvement from previous quarters and demonstrates the growing capability of the ecosystem to respond to and mitigate the impact of attacks. Additionally, despite Ethereum’s TVL growing by nearly 400% year-on-year, it only suffered $8 million in losses this quarter, indicating some improvement in DeFi defenses. This resilience in the face of rapid growth is an encouraging sign for the industry.
Credit: Tesfu Assefa
The Importance of Audits
The reports reveal a critical gap in security practices among many projects. Out of 41 hacked projects analyzed, only seven had undergone the relevant audits. This alarming statistic underscores the vital importance of thorough security measures in preventing large-scale exploits, including regular audits and robust bug bounty programs.
History has shown that projects that prioritize these security measures are less likely to fall victim to attacks, signposting a clear path to better security.
Conclusion
As we move forward, a collaborative effort between developers, security researchers, and users will be crucial in building a more resilient and secure crypto ecosystem. The industry must prioritize security measures to protect users and maintain trust. By learning from these incidents, implementing stronger security protocols, and fostering a culture of vigilance, the crypto ecosystem can work towards a more secure future for all participants.
Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter.
It’s standard procedure that cryptocurrency projects come and go at a dizzying rate, as they often serve no real immediate purpose. However, some protocols have managed to establish themselves as revenue-generating powerhouses, demonstrating real-world utility, user adoption, and sustainable profits.
Traditional Finance firms are chomping at the bits for the newly-approved Ethereum spot ETF to start trading. The Bitcoin ETF serves as a safe haven asset hedge, ETH is an asset class that Wall Street can engage with. TradFi firms can use metrics like new users, fees, revenue and total value locked (TVL) to measure network effect. With Ethereum clearing the way, other chains and protocols can eventually follow in its wake.
We’ve used a recent study by Onchain Times and Token Terminal data to do a deep analysis of the top ten money spinners in crypto in mid-2024, comparing their business models, revenue streams, and key performance metrics.
1. Ethereum: The Undisputed Leader
Ethereum remains the giant of the crypto industry, generating an impressive $1.42 billion in revenue year-to-date (YTD). As the foundation for much of the decentralized finance (DeFi) ecosystem, Ethereum’s success stems from its widespread adoption and the high demand for block space on its network, as well as recent upgrades like the Merge and Proto Danksharding upgrade, which has moved it to proof-of-stake and slashed layer-2 chain costs.
Key points
Highest revenue generator in the crypto space
Revenue primarily comes from transaction fees paid by users
Profitability fluctuates due to issuance rewards to validators
Q1 2024 was profitable, while Q2 saw a decline due to activity moving to layer-2 solutions
2. Tron: The Stablecoin Highway
Surprising many, Tron takes the second spot with approximately $852 million in revenue YTD (year to date). Tron’s success is largely attributed to its role as a major conduit for stablecoin transfers, particularly USDT in developing economies. It’s cheap, fast, and reliable.
Key points
Second-largest stablecoin ecosystem after Ethereum
Popular in countries like Argentina, Turkey, and various African nations
Competes with Ethereum and Solana for highest stablecoin transfer volumes
3. Maker: The OG Stablecoin Protocol
Maker, the protocol behind the DAI stablecoin, comes in third with $176 million in revenue YTD. Its business model revolves around issuing DAI against crypto collateral and charging interest on these loans.
Key points
Total DAI supply is currently 5.2 billion, down from its all-time high of around 10 billion
It has diversified revenue streams, including holding real-world assets (RWA) at 25.6% of total revenue
Estimated earnings of $73 million annually after accounting for DAI Savings Rate and operating costs
4. Solana: The Phoenix Rising (Again)
Once written off as dead, Solana has made an impressive comeback since its 2023 Breakpoint conference, ranking fourth with $135 million in annualized revenues YTD. Its resurgence is attributed to increased activity in memecoins, NFTs, and DePIN (Decentralized Physical Infrastructure Networks) projects.
Key points
Revenue comes from transaction fees paid to validators
High token issuance costs make it challenging to assess profitability
Success driven by technological improvements and community-driven events like the JTO airdrop
5. Ethena: The New Stablecoin Contender
Launched in January 2024, Ethena has quickly become the fifth-largest revenue-generating protocol, with $93 million in annualized revenues. It’s backed by big names like Arthur Hayes, and while it’s conjured up some early Luna 2.0 fears due to its algorithmic stablecoin design, so far it’s doing well. Its USDe token, a synthetic dollar, has achieved a market cap of $3.6 billion in just a few months.
Key points
Innovative delta hedging strategy to maintain USDe peg
Business model designed to excel in bull markets, raising questions about long-term sustainability
6. Aerodrome: The Base Layer AMM
Aerodrome, an automated market maker (AMM) on the Base layer-2 network, has generated $85 million in revenue YTD. Launched in August 2023, it has quickly established itself as the top decentralized exchange (DEX) on Base.
Key points
Implements successful mechanisms from various DEX protocols
Uses vote-escrowed tokenomics to attract liquidity
Incorporates concentrated liquidity features to compete with Uniswap
7. Lido: The Liquid Staking Giant
Lido, a prominent liquid staking protocol, has generated $59 million in revenue year-to-date across Ethereum and Polygon proof-of-stake chains. Its popularity stems from making Ethereum staking more accessible to average users.
Key points
Revenue comes from a 10% fee on users’ staking rewards
Profits of $22.5 million YTD after accounting for node operator payments and token incentives
Operates as a double-sided market, connecting ETH holders with professional node operators
8. Base: The Coinbase L2 Solution
Base, a fast-growing Ethereum layer-2 solution launched by Coinbase in Q3 2023, clocks in at $52 million in revenues YTD. As a relatively new entrant, its rapid growth is noteworthy, and its backing by Coinbase could see it reach the top of the food chain very quickly.
Key points
Revenue comes from user transaction fees
Impressive profitability with $35 million in earnings YTD
Benefited significantly from the implementation of EIP-4844 that reduced data availability costs
9. Uniswap Labs: The DEX Pioneer
Uniswap Labs, the company behind the popular decentralized exchange Uniswap, has generated $39.3 million in revenue YTD. Uniswap was the earliest DEX to gain real traction, and continues to play a crucial role in the DeFi ecosystem.
Key points
Revenue primarily comes from trading fees
Pioneered the automated market maker (AMM) model in DeFi
Continues to innovate, with features like concentrated liquidity in Uniswap V3
10. PancakeSwap: The BSC DeFi Leader
PancakeSwap, a leading DEX on the Binance Smart Chain (BSC), rounds out the top ten revenue-generators, with $36.3 million in revenue YTD. Its success highlights the growing importance of alternative blockchain ecosystems.
Key points
Largest DEX on Binance Smart Chain
Offers a wide range of DeFi services – including trading, yield farming, and NFTs
Lower transaction costs compared to Ethereum-based DEXs
Credit: Tesfu Assefa
Comparing the Ten Chains:
Revenue Generation (year-to-date)
Ethereum: $1.42 billion
Tron: $852 million
Maker: $176 million
Solana: $135 million
Ethena: $93 million
Aerodrome: $85 million
Lido: $59 million
Base: $52 million
Uniswap Labs: $39 million
PancakeSwap: $36 million
Ethereum’s revenue still dwarfs that of its competitors, emphasizing its dominant position. However, the presence of new entrants like Ethena, Base, and established DEXs like Uniswap and PancakeSwap shows that revenue is chain-agnostic and that investors will find it wherever they can.
Remember the importance of understanding tokenomics; Lido, for example, still trades at under $2, the same price it had two years ago, despite its market cap growing 50x. When assessing a cryptocurrency, look at its fully diluted value (FDV) instead of current market cap.
Profitability
Profitability varies significantly among these protocols due to differences in their business models and their running cost:
Ethena: leads in profitability with $41 million in earnings YTD.
Base: shows strong profitability with $35 million in earnings.
Maker: estimates $73 million in annualized earnings after costs.
Lido: reports $22.5 million in profits YTD.
Ethereum and Solana’s profitability is more complex due to token-issuance costs.
Profitability data for Uniswap Labs and PancakeSwap is not readily available.
Business Model Diversity
The top cash cows in crypto have diverse business models:
Infrastructure providers: Ethereum, Tron, Solana, Base
There is more than one way to skin a cat. Protocols in the crypto ecosystem can generate revenue in entirely different ways – from providing foundational infrastructure to offering specific financial services.
Market Position and Competition
Ethereum maintains its leadership position, but faces growing competition from layer-2 solutions and alternative layer-1 blockchains.
Tron has carved out a niche in stablecoin transfers, particularly in developing markets.
Maker continues to be a major player in the stablecoin space, but faces new competition from innovative protocols like Ethena.
Solana has shown resilience and adaptability, rebounding from near-collapse to generate healthy revenue.
Base and Aerodrome demonstrate the potential for new entrants to quickly gain market share with innovative features and strong backing.
Uniswap and PancakeSwap showcase the ongoing importance of decentralized exchanges, with each dominating their respective blockchains.
Sustainability and Future Outlook
When assessing these protocols, it’s crucial to consider the sustainability of their revenue models:
Ethereum’s shift to proof-of-stake and the growth of layer-2 solutions may impact its long-term revenue structure.
Tron’s reliance on stablecoin transfers could be vulnerable to regulatory changes or shifts in market dynamics.
Maker’s diversification into real-world assets may provide more stable revenue streams.
Ethena’s success in bull markets raises questions about its performance during market downturns.
Base and Aerodrome will need to maintain their innovative edge to continue attracting users and liquidity.
Uniswap and PancakeSwap face increasing competition from other DEXs, and may need to continue innovating to maintain their position in a competitive market.
Conclusion
The top ten cash cows in crypto are a mix of established giants, innovative newcomers, and specialized DeFi protocols. While Ethereum continues to dominate in terms of raw revenue, the success of newer protocols like Ethena and Base, as well as the continued relevance of DEXs like Uniswap and PancakeSwap, demonstrates the ongoing evolution and diversification of the crypto landscape.
The presence of both infrastructure providers and application-layer protocols in this list highlights the importance of a robust and diverse ecosystem. Investors and users should closely monitor these protocols, as their performance often serves as a barometer for broader trends in crypto.
Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter.
In the realm of programming languages and formal methods, the representation and manipulation of syntax, particularly for languages with complex variable binding structures, is a significant challenge. Traditional methods often involve cumbersome and error-prone techniques, such as manually handling variable binding and substitution. However, recent advancements have introduced more robust and systematic approaches. One such advancement is presented in a recent study, which outlines a framework for automatically generating Agda implementations of second-order languages. This article explores the main concepts of this framework, its foundations, and its implications for the field.
Understanding the Framework
At its core, the framework allows users to produce implementations of second-order languages in Agda with minimal manual effort. The generated term language is explicitly represented as an inductive, intrinsically-encoded data type. This means that the structure and rules of the language are built directly into the data type definitions, ensuring that terms are always well-formed according to the language’s syntax and semantics.
This intrinsic encoding offers several advantages over traditional approaches. By embedding the rules directly into the data type definitions, the framework ensures that any term constructed is guaranteed to be syntactically correct. This reduces the likelihood of errors and simplifies the reasoning about programs and their properties.
The framework supports various formalised metatheoretical constructs, such as substitution for operational semantics and compositional interpretations for denotational semantics. These constructs are essential for defining how the language behaves and how terms can be transformed and interpreted. For example, substitution is crucial for operational semantics, defining how variables in a program can be replaced with their corresponding values. Compositional interpretations, on the other hand, are key for denotational semantics, allowing for a systematic interpretation of programs in a mathematical domain.
Mathematical Foundations
The framework’s strength lies in its deep mathematical foundations, specifically derived from the theory of abstract syntax. Traditional approaches often require ad-hoc definitions and lemmas to handle variable binding and substitution, leading to complex and error-prone implementations. In contrast, the presented framework leverages a systematic mathematical approach, avoiding these pitfalls.
One significant mathematical tool used in this framework is the presheaf model. This model provides a structured way to handle variable binding by treating contexts (environments in which variables exist) as functors. This approach allows for a more elegant and powerful handling of variable scopes and substitutions, which are crucial for both the correctness and usability of the language representations.
Presheaves provide a categorical framework that simplifies many of the complexities associated with variable binding. They allow for the definition of substitution and other operations in a way that is both mathematically rigorous and practically useful. By treating contexts as functors, the framework can systematically handle variable scopes and avoid common pitfalls such as variable capture and name clashes.
Related Work and Comparisons
The challenge of formalising and reasoning about abstract syntax has a rich history, motivated largely by the development of proof assistants. The Barendregt variable convention, which suggests renaming variables to avoid clashes, is notoriously difficult to formalise. Several approaches have been developed to tackle this issue, including higher-order abstract syntax, locally nameless representation, and intrinsically-typed encoding.
Higher-order abstract syntax, introduced by Pfenning and Elliot, represents variables and bindings using the meta-language’s own functions and variables. This approach simplifies many aspects of the implementation but can be less efficient for certain operations. For example, while higher-order abstract syntax can make it easier to define certain operations, it can also introduce inefficiencies when manipulating large terms or performing complex substitutions.
Locally nameless representation, as explored by Bird and Paterson, uses a hybrid approach, combining named and nameless (de Bruijn indices) representations to balance ease of use and efficiency. This approach allows for more efficient manipulation of terms while still providing a systematic way to handle variable binding. However, it can still be prone to errors and require complex arithmetic operations.
Intrinsically-typed encoding, as employed in the discussed framework, ensures that terms are always well-typed by construction. This method avoids many of the pitfalls of other approaches, such as the complicated arithmetic involved in de Bruijn indices. By embedding the typing rules directly into the data type definitions, intrinsically-typed encoding provides strong guarantees about the correctness of terms and simplifies the reasoning about programs.
Advantages of the Presented Framework
The framework’s approach to intrinsically-typed representation offers several advantages. First, it provides strong static guarantees about the typing and scoping of terms, reducing the risk of errors. This is particularly valuable in dependently-typed proof assistants like Agda, where correctness proofs are central. By ensuring that terms are always well-typed, the framework simplifies the development and verification of programs and reduces the likelihood of errors.
Moreover, the framework includes a code-generation script that facilitates rapid prototyping and experimentation. This script allows users to quickly generate and test new language features or modifications, significantly speeding up the development process. For example, a researcher can easily define a new language construct, generate the corresponding Agda implementation, and immediately begin experimenting with its properties and behaviour.
Another noteworthy feature is the framework’s ability to incorporate generic traversals and equational logic through parameterized meta variables. This capability simplifies the manipulation and reasoning about terms, making it easier to develop complex language features and proofs. For example, the framework can automatically generate code for performing common operations, such as substitution or evaluation, and provide systematic ways to reason about their correctness.
Case Studies and Benchmarks
The framework was evaluated using the PoplMaRK challenge, a set of benchmarks for comparing metatheory formalisation efforts. Many existing approaches, particularly those using Coq, rely on numeric de Bruijn indices, which can be complex and error-prone. In contrast, the presented framework’s use of an intrinsically-typed, nameless representation proved more robust and easier to manage.
The PoplMaRK challenge includes a variety of tasks designed to test the capabilities of different formalisation frameworks. These tasks range from simple operations, such as substitution and evaluation, to more complex ones, such as proving properties about the language and its semantics. By demonstrating the framework’s ability to handle these tasks efficiently and correctly, the authors provided strong evidence of its robustness and utility.
Credit: Tesfu Assefa
Future Directions
The framework’s creators recognize that modern type theory encompasses a wide range of formal systems beyond second-order languages with algebraic types. Future work aims to extend the framework to handle these more complex systems, such as linear, dual-context, polymorphic, dependent, and polarised calculi. This expansion would further enhance the framework’s utility and applicability.
Additionally, ongoing work focuses on refining the categorical reformulation of the presheaf model to suit the practical needs of formalisation. This involves developing new notions and techniques to avoid quotienting, making the formalisation process more efficient and user-friendly. By addressing these challenges, the authors hope to further simplify the development and verification of complex language systems.
The framework’s flexibility and extensibility make it well-suited for a variety of applications. For example, it could be used to formalise and verify the semantics of new programming languages, develop tools for program analysis and optimization, or even explore new mathematical theories related to syntax and semantics. As the field continues to evolve, the framework’s capabilities will likely expand, enabling researchers to tackle increasingly complex problems.
Conclusion
The framework for generating Agda implementations of second-order languages represents a significant advancement in the field of programming languages and formal methods. By leveraging deep mathematical foundations and providing robust, systematic tools, this framework simplifies the development and verification of complex language systems. Its intrinsic typing guarantees, ease of extension, and support for rapid prototyping make it a valuable asset for researchers and developers alike.
As the field continues to evolve, the principles and techniques introduced by this framework will likely inspire further innovations, driving progress in the formalisation and implementation of increasingly sophisticated language systems. The future work outlined by the framework’s creators promises to expand its capabilities, addressing more complex and varied language constructs, and further solidifying its place as a cornerstone in the study of programming languages and formal methods.
In summary, this framework provides a powerful and flexible tool for the formalisation of second-order languages, offering significant improvements over traditional approaches. Its mathematical rigour, combined with practical tools for rapid development and experimentation, makes it an invaluable resource for both researchers and practitioners. As we look to the future, the framework’s potential for further development and application promises to drive continued progress in the field, opening up new possibilities for the study and implementation of programming languages.
Reference
Fiore, Marcelo, and Dmitrij Szamozvancev. “Formal Metatheory of Second-order Abstract Syntax.” Proceedings of the ACM on Programming Languages 6, no. POPL (January 12, 2022): 1–29. https://doi.org/10.1145/3498715.
Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter.
The cryptocurrency sector recently witnessed an intriguing security event in the Cardano ecosystem – a distributed denial-of-service (DDoS) attack that was swiftly mitigated, showing the robustness of the blockchain and the ingenuity of its developer community, marking a victory for the collaborative spirit that defines the crypto space.
Let’s delve into the details of this attack, its resolution, and the implications for Cardano’s future.
What is a DDoS attack?
DDoS (distributed denial of service) and DoS (denial of service) attacks have been a thorn in the side of Web2 businesses since 1996, causing billions of dollars of losses in the process. In the crypto world, there haven’t been many, although Electrum Wallet’s 2019 incident is a notable one: created a botnet of 152,000 hijacked wallets, and in 2020, two exchanges were shut down by DDoS attacks.
In short, DDoS (Distributed Denial of Service) attacks are malicious attempts to disrupt the normal operation of a cryptocurrency network, exchange, or service. Attackers flood the target with a massive amount of internet traffic from multiple sources, overwhelming its infrastructure and causing it to become unavailable to legitimate users. These attacks can have serious consequences: they can halt trading, block transactions, and cause financial assets to be lost.
Cardano DDoS Attack: Play-by-Play
On 24 June, 2024, the Cardano network experienced an unusual surge in activity. Fluid Token’s CTO reported that the attack commenced at block 10,487,530.
The attacker’s strategy was to flood the network with transactions, each executing 194 smart contracts. At a cost of 0.9 ADA ($0.36) per transaction, the malicious actor attempted to congest the network by filling blocks with these spam transactions.
The intent behind this DDoS attack was twofold:
Primarily, it aimed to disrupt the network’s normal operations by overwhelming it with traffic.
There was speculation that the attacker might have also been attempting to manipulate the fee structure to enable cheaper high-value transactions, or maybe to steal staked ADA tokens.
The attack resulted in a significant increase in network load, with chain utilization reaching peaks of 72% on average and up to 93% on an hourly basis. This heightened activity raised concerns among community members and developers who noticed the network’s sluggish performance.
Community Response and Investigation
As news of the attack spread, the Cardano community quickly mobilized. Developers, led by figures such as Philip Disarro from Anastasia Labs, began investigating the attack and formulating countermeasures.
Through on-chain analysis, community members traced the origin of the attack to addresses linked to the Kraken exchange. This discovery raised questions about the attacker’s identity and the potential for legal action. The transparency of blockchain technology was invaluable in this investigation, allowing for real-time tracking of the malicious transactions.
Interestingly, despite the attacker’s efforts to congest the network, their actions inadvertently contributed over 1,000 ADA in transaction fees to the Cardano treasury and stake pool operators. This unintended consequence showed how the Cardano network’s economic model can help keep it safe.
Technical Analysis and Vulnerability Discovery
As the community rallied to understand and counter the attack, developers like Mel from Harmonic Labs began dissecting the malicious transactions. By deserializing the UPLC (Untyped Plutus Core) of the attacking scripts, they discovered a critical flaw in the attacker’s strategy.
The scripts used in the attack were designed to always return ‘true’, no matter what input they were fed. This oversight meant that the scripts could be easily manipulated, providing an opportunity for the defenders to turn the tables on the attacker.
The Counterattack: A Brilliant Solution
Philip Disarro of Anastasia Labs identified a clever way to not only stop the attack but also claim the attacker’s funds. The solution involved deregistering the stake credentials used by the attacker. This action would force the attacker to re-register their credentials at a cost of 400 ADA each time they wanted to continue the attack, significantly increasing the financial burden of their malicious activities.
Moreover, this countermeasure allowed defenders to claim the attacker’s ADA, effectively turning the attack into a donation to the Cardano ecosystem.
As Disarro put it:
Thanks for the free money, moron.
The community quickly implemented this solution, deregistering approximately 200 stake contracts from the attacker, which did the trick.
Credit: Tesfu Assefa
Lessons Learned and Network Resilience
The failed DDoS attack provided several valuable insights into Cardano’s capabilities:
1. Network Capacity: Despite the high transaction volume, Cardano’s network continued to function, processing legitimate transactions alongside the spam. This demonstrated the blockchain’s ability to handle significantly increased loads, suggesting room for future scaling.
2. Community Strength: The rapid response and clever solution showcased the strength of Cardano’s developer community. Their ability to quickly analyze, respond, and implement countermeasures highlights the importance of a robust and engaged team.
3. Economic Model: The attack inadvertently proved the effectiveness of Cardano’s economic model. The attacker’s funds were not only used to pay transaction fees, but were also claimed by the defenders, turning a potential threat into a net positive for Cardano.
4. Transparency: The ability to track and analyze the attack in real-time demonstrated the value of blockchain transparency in security and incident response.
Future Implications and Upgrades
In the aftermath of the attack, the Cardano development team, including organizations like Intersect, began working on node upgrades to bolster the network’s resilience against these attacks. The upgrades aim to address potential vulnerabilities without compromising the network’s performance or decentralization.
The incident also sparked discussions about potential parameter adjustments, such as increasing block sizes or reducing block times, to further improve the network’s capacity and resilience.
Comparison with Other Networks
This event provided an interesting contrast to how other blockchain networks handle similar attacks. As noted in the community discussions, when Solana faces attacks, it often results in network shutdowns. Ethereum, on the other hand, typically sees transaction fees skyrocket during periods of network congestion.
Cardano’s ability to withstand the attack with only mild degradation in performance, coupled with the community’s innovative response, proves it is a robust and resilient blockchain platform.
Conclusion
The recent DDoS attack on Cardano, while potentially disruptive, ultimately served to demonstrate the strength and resilience of the network and its community. The swift and clever response thwarted the attack – and even turned it into an opportunity for growth and improvement. While Cardano has had its share of criticism – including some undeserved ridicule – for its slow development, its security has now been battle-tested and is hard to criticize.
As Cardano continues to evolve, incidents like these provide valuable lessons and drive innovation. They underscore the importance of ongoing development, community engagement, and the power of decentralized systems in facing security threats.
The crypto world will undoubtedly be watching Cardano’s continued development with interest, as it sets new standards for blockchain resilience and community-driven problem-solving.
Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter.
On Saturday, July 13, 2024, former U.S. President Donald Trump was the target of a shocking assassination attempt during a campaign rally in Pennsylvania. The incident, which saw Trump narrowly escape death, has sent shockwaves through the political landscape and financial markets, including the cryptocurrency sector, with Bitcoin and friends immediately surging in price in its wake.
This article examines the events surrounding the assassination attempt and its subsequent immediate and long term impact on the crypto markets.
The Trump Assassination Attempt
During a campaign event in Butler, Pennsylvania, a gunman attempted to assassinate Donald Trump, the Republican nominee for the 2024 U.S. presidential election as he was addressing supporters. Trump was shot in the ear but survived the attack and was ushered off the stage waving a defiant fist in the air, which was captured by a photographer for an iconic image. His campaign reported that he was doing well following the incident.
Public sentiment has shifted dramatically in favor of Trump since the incident. He is now the 60% favorite according to betting markets in the presidential election in November.
Immediate Market Reaction
In the aftermath of the assassination attempt, cryptocurrency markets experienced a significant surge, after weeks of decline due several factors, including summer holidays, bearish market pressure caused by Germany selling seized Bitcoin, as well the announcement that Mt Gox would begin returning stolen BTC to victims of the 2014 hack.
After dropping as low as $53k during early July, Bitcoin (BTC), the world’s leading cryptocurrency and the one that dictates the overall market confidence in digital assets, saw a sharp increase in value, rising by more than 25% to reach $66.4k by Friday, July 19. This marked its highest level in four weeks and represented a year-to-date gain of approximately 54%. Other cryptocurrencies also benefited from the market movement, with Ethereum (ETH) rising 12.1% to $3,488.
The surge in crypto prices was not isolated to major currencies. Meme tokens associated with Trump also experienced substantial gains. For instance, MAGA leapt from around $6.35 to a brief peak over $9.50, the satirical TREMP token is up 15$ this week. Conversely, BODEN, a joke asset named after President Joe Biden, has declined by about 18%.
Hundreds of pro-Trump memecoins were launched in the days after the shooting, most pumping and dumping within hours.
5 Reasons For The Trump Shooting Crypto Rally
Several factors contributed to the cryptocurrency market’s positive reaction to the assassination attempt:
1. Increased Trump Victory Odds
The incident appears to have bolstered Trump’s chances of winning the presidency. Betting markets and political analysts suggest that the attack may garner sympathy votes and mobilize his base to vote. On the Polymarket prediction platform, the probability of Trump winning the presidency jumped to an all-time high of 70% following the incident.
2. Trump’s Pro-Crypto Stance
After slamming crypto during his first term, Trump has made a remarkable recent U-turn on crypto after it became clear that the tens of millions of Americans owning crypto could ultimately decide the next president. As a result, Joe Biden flip flopped soon after to also embrace crypto and the SEC stunningly approved a spot Ethereum ETF.
Throughout his campaign so far, Trump has positioned himself as a champion of cryptocurrency. He has hosted cryptocurrency industry executives at Mar-a-Lago, and expressed enthusiasm for Bitcoin mining in the USA. Trump’s campaign is also the first from a major U.S. political party to accept cryptocurrency payments, signaling a potential shift in the regulatory landscape if he were to win the election.
3. Anticipated Regulatory Changes
Investors speculate that a Trump presidency could lead to a more favorable regulatory environment for cryptocurrencies. Trump has criticized Democratic attempts to regulate the crypto sector through the SEC and the controversial Operation Chokepoint 2.0 which has seen an exodus of Web3 firms from the USA, leading many to believe his administration would adopt a lighter touch approach to oversight.
4. Economic Policy Expectations
Trump’s previous tenure was marked by tax cuts and deregulation. Investors anticipate similar policies in a potential second term, which could drive up deficits and inflation. Such economic conditions often lead investors to seek alternative assets like cryptocurrencies as a hedge.
5. Trump’s VP pick is pro-crypto
The crypto market rally gained further momentum on Monday, July 15, when Trump announced 39 year-old Senator J.D. Vance of Ohio as his running mate. Vance, known for his tech-savvy background and pro-crypto stance, has further energized the digital asset community. He previously declared $250k in Bitcoin holdings and is an outspoken critic of crypto’s arch-enemy, SEC chairman Gary Gensler.
Vance’s selection is seen as a strategic move to appeal to both traditional conservatives and the tech-oriented younger demographic.
The Trump-Vance Ticket and What It Means for Crypto
Vance’s pro-crypto credentials include:
Personal investment in Bitcoin through Coinbase, valued between $100,001 and $250,000
Support for bills promoting cryptocurrency innovation in the Senate
Opposing increased regulatory scrutiny of the crypto industry
Drafting legislation to overhaul digital asset regulation
The Trump-Vance ticket is viewed favorably by many in the crypto and tech industries. Notable figures – such as Peter Thiel (a mentor of Vance), Marc Andreessen (who has now also endorsed Trump publicly), Ben Horowitz, and the Winklevoss twins – have expressed support for the candidates. This backing from influential tech personalities adds credibility to the ticket’s pro-crypto stance.
Credit: Tesfu Assefa
USA and Crypto: Potential Trump Policy Shifts
Experts anticipate several policy changes under a potential Trump-Vance administration that could benefit the cryptocurrency market:
1. Deregulation: A return to Trump’s previous deregulatory approach could create a more favorable environment for crypto entrepreneurs and investors.
2. Currency Devaluation: Vance has advocated for a weaker dollar, which could indirectly boost Bitcoin’s value proposition as a hedge against currency devaluation.
3. Crypto-Friendly Banking Reforms: Policies making it easier for traditional institutions to hold their clients crypto in custody could lead to broader adoption.
4. Redefining Crypto Assets: A potential shift in how cryptocurrencies are classified could impact their regulatory treatment. At present only Bitcoin and now Ethereum have been greenlit as non-securities.
5. Antitrust Actions Against Big Tech: Vance supports antitrust measures against major tech companies, and this could create opportunities for blockchain-based Web3 alternatives, which hold huge potential to disrupt the monopoly of Web2 giants.
Market Outlook and Expert Opinions
While the crypto market has responded positively to these political developments, opinions vary on the long-term implications:
Matthew Sigel, head of digital assets research at VanEck, believes the Trump-Vance ticket represents a significant shift toward a more crypto-friendly regulatory environment. He suggests their pro-business stance could pave the way for a more favorable environment for crypto entrepreneurs and investors.
Mark Cuban, billionaire entrepreneur and investor, posited on social media in a long post that the support from Silicon Valley for the Trump-Vance ticket might be motivated by potential benefits to Bitcoin. He argues that lower tax rates and tariffs, combined with global uncertainty about the USA’s geopolitical role, could accelerate Bitcoin’s price and potentially establish Bitcoin as a global ’safe haven’ currency.
Conclusion
The assassination attempt on Donald Trump and the subsequent selection of J.D. Vance as his running mate have provided an unlikely shot-in-the-arm for the cryptocurrency markets. The rally in Bitcoin and other digital assets reflects renewed investor optimism about the future of the space, which may soon be under more crypto-friendly regulations.
There is a near-consensus opinion that the Fed will reduce interest rates by a quarter of a percentage point, or 25 basis points, for the first time since the end of 2021, so the stars seem to be aligning for Bitcoin and its children.
Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter.
Medical image segmentation is a critical task in healthcare, providing essential information for diagnosis and treatment planning. Traditional methods in this domain often suffer from significant limitations, including the need for retraining models for new tasks and the production of deterministic outputs that fail to capture medical image ambiguities. Researchers at MIT have made a substantial leap forward with the introduction of Tyche, a novel framework that leverages stochastic in-context learning to address these challenges.
Addressing Core Challenges in Medical Image Segmentation
The segmentation of medical images involves identifying and delineating structures within various imaging modalities like MRI, CT scans, and ultrasounds. Existing models typically require retraining for new segmentation tasks, a process demanding considerable computational resources and domain expertise. Moreover, these models generally produce a single, deterministic result, which does not account for the inherent variability and ambiguity in medical images.
Introducing Tyche: An Innovative Framework
Tyche is designed to overcome these limitations by integrating probabilistic segmentation with an in-context learning framework. The framework operates with two primary variants:
Tyche-Train-time Stochasticity (Tyche-TS): This variant is trained to produce multiple segmentation candidates during the training phase. It learns the distribution of possible labels and generates diverse segmentations by allowing interactions among different predictions.
Tyche-Inference-time Stochasticity (Tyche-IS): This utilizes a deterministic model trained traditionally and employs test-time augmentation during inference to produce varied segmentations without the need for retraining.
Credit: Tesfu Assefa
Methodology and Technical Innovations
The Tyche framework requires an image to be segmented and a contextual set of image-segmentation pairs that define the task. Tyche-TS incorporates stochastic elements during training to encourage diversity in segmentation predictions. This is achieved using the innovative SetBlock mechanism, which integrates multiple predictions and introduces noise to foster diverse candidate outputs. In contrast, Tyche-IS generates multiple segmentation candidates during inference by applying augmentation techniques to both the target image and the context set. This approach effectively diversifies predictions using a pre-existing deterministic model, ensuring robust performance without additional training.
A key component of Tyche’s methodology is the Best Candidate Dice Loss, tailored to optimize the best prediction among the multiple candidates. This loss function drives the model towards generating a variety of plausible segmentations, thereby enhancing the flexibility and applicability of the framework.
Demonstrated Efficacy and Versatility
The researchers tested Tyche on twenty unseen medical imaging tasks, benchmarking its performance against existing methods. The results were impressive, with Tyche displaying superior performance when compared to both in-context learning baselines and interactive segmentation methods. Notably, Tyche’s results closely aligned with those produced by specialized stochastic models meticulously trained for specific tasks.
One of Tyche’s most significant advantages is its generalizability. The framework effectively handles images from datasets not encountered during training, outperforming other segmentation methods regarding metrics like the best candidate Dice score and Generalized Energy Distance (GED). This capability is particularly valuable in clinical settings where the diversity of tasks and image types is vast.
Practicality and Efficiency
Tyche’s two variants offer a well-balanced trade-off between computational efficiency and prediction quality. Both Tyche-TS and Tyche-IS have shown to be practical for clinical use, making them viable options for real-world medical imaging applications.
https://arxiv.org/abs/2401.13650Figure 1: Visualization of predictions for three different samples, 1 per row. Left: LIDC-IDRI. Right: Hippocampus dataset. The leftmost columns are raters’ annotations. The 4 last columns are model predictions. Tyche provides a set of prediction that is diverse and matches the raters, for tasks unseen at training time. (Credit: Rakic et al., “Tyche: Stochastic In-Context Learning for Medical Image Segmentation.”)
Future Directions
The advent of Tyche opens several exciting avenues for future research. Further exploration is needed to understand the types of uncertainty captured by its stochastic mechanisms. Additionally, expanding Tyche’s applicability to more complex support sets and various image modalities could broaden its impact significantly.
Conclusion
Tyche represents a transformative advancement in the field of medical image segmentation. By addressing the need for stochastic predictions and reducing the dependency on task-specific retraining, Tyche delivers robust, diverse, and practical solutions for medical professionals. Its ability to generalise across different tasks and datasets paves the way for more efficient and effective medical imaging applications, heralding a new era in medical diagnostics and treatment planning.
Reference
Rakic, Marianne, Hallee E. Wong, Jose Javier Gonzalez Ortiz, Beth Cimini, John Guttag, and Adrian Dalca V. “Tyche: Stochastic In-Context Learning for Medical Image Segmentation.” arXiv.org, January 24, 2024. https://arxiv.org/abs/2401.13650.
.png)
.png)
.png)