Topics

Follow the subjects shaping intelligence, science, culture, and the future of technology.

Browse topics

Find a subject, then follow its latest reporting and ideas.

Neural Network

Stories, interviews, and analysis in this subject.

Conscious Only While Learning? Continual Learning, Embodiment, and Our Mind Children
Article Nov 6, 2026

Conscious Only While Learning? Continual Learning, Embodiment, and Our Mind Children

From frozen models to mind children: learning, bodies, and experience. Continual learning, embodiment, and why frozen AI weights may not settle the question of machine consciousness.
Giulio Prisco 10 min read
A language model that can read and write neural network weights
News Nov 3, 2026

A language model that can read and write neural network weights

MIT CSAIL researchers show that a frontier language model can understand, repair, and even create neural network weights directly, hinting at a new way to optimize and interpret AI systems.
Giulio Prisco 2 min read
Brainwaves That Say 'Not What I Meant': AI Learns to Read Unspoken Corrections
News Oct 12, 2026

Brainwaves That Say 'Not What I Meant': AI Learns to Read Unspoken Corrections

Researchers at KAIST and Microsoft Research Asia developed Neural Value Alignment, a system that reads EEG signals in real time to detect when AI misunderstands human intent. By distinguishing between reward prediction error and state prediction error, the technology enables machines to self-correct without explicit verbal feedback, pointing toward more natural human-AI collaboration.
Giulio Prisco 2 min read
Bio-Inspired Adaptive Neurons For Dynamic Weighting In Artificial Neural Networks
Community Apr 6, 2026

Bio-Inspired Adaptive Neurons For Dynamic Weighting In Artificial Neural Networks

Bio-inspired adaptive neurons that dynamically adjust weights during inference, bringing biological flexibility to rigid artificial neural networks for smarter, more adaptable AI.

Hermela Wesene 4 min read
Introducing Temporal Kolmogorov-Arnold Networks (TKANs): A Breakthrough in Multi-step Time Series Forecasting
Community Apr 26, 2025

Introducing Temporal Kolmogorov-Arnold Networks (TKANs): A Breakthrough in Multi-step Time Series Forecasting

TKANs reshape prediction: faster, sharper, and easier to trust—built on timeless theorems, not just bigger networks.

Deborah Gulma 6 min read
Text-to-LoRA: Instant Transformer Adaptation
Community Dec 30, 2025

Text-to-LoRA: Instant Transformer Adaptation

Discover how Text-to-LoRA (T2L) enables instant, zero-shot adaptation of LLMs—compressing hundreds of LoRA adapters, generating task-specific weights from language, and personalizing AI behavior effortlessly.

Peter Kinfe 4 min read
The Hidden Shapes Inside Every Network
Community Apr 6, 2026

The Hidden Shapes Inside Every Network

Scientists just cracked a decades-old problem: SPMiner uses Graph Neural Networks to discover hidden recurring shapes inside massive networks faster and at unprecedented scale.

Ruth Tamiru 5 min read
Unlocking Neural Network Potential: The Power of Random Weights and Learned Biases
Article May 10, 2025

Unlocking Neural Network Potential: The Power of Random Weights and Learned Biases

Neural Networks learn without Weight updates? New study shows Bias-Only training matches full models. A minimalist approach to training challenges deep learning orthodoxy, offering a lighter path to high-functioning models.

Bitania Tilahun 8 min read
Unveiling the Hidden Dynamics: How Neural Networks Navigate Latent Spaces
Community Dec 30, 2025

Unveiling the Hidden Dynamics: How Neural Networks Navigate Latent Spaces

Discover how autoencoders reveal hidden neural dynamics: latent vector fields, attractors, and trajectories expose memorization, generalization, and data-free insights—reshaping interpretability, OOD detection, and AI understanding.

Simon Assfaw 3 min read
Building Better Brains: The NeuroEvolution of Augmenting Topologies
Community May 22, 2026

Building Better Brains: The NeuroEvolution of Augmenting Topologies

NEAT evolves neural networks like living organisms, starting simple, protecting innovation, and growing smarter architectures that learn dramatically faster through artificial evolution.

Kaleab Hailemeskel 4 min read