Science ❯ Computer Science ❯ Artificial Intelligence

Neural Networks

Research Geoffrey Hinton John Hopfield Boltzmann Machine Prof Geoffrey Hinton Prof John Hopfield Hopfield Network Foundational Models Research Contributions Hopfield Networks Ethics in AI Physics-Informed Neural Networks Machine Learning Impact of AI on Society Model Training Parameters Image Classification Common Features Backpropagation Feature Representation Image Processing Neural-Symbolic Systems Optimization Techniques Generalization Issues Scaling Techniques Signal Processing Stability in Estimation Model Compression Frequency Selection Representation Learning Mixture-of-Experts Weight Updating Activation Distribution Feature Learning Model Behavior Coalitional Game Theory Computer Vision Attention Mechanisms Implicit Neural Representation Regularization Techniques Implicit Neural Representations DeepMind LoRA Techniques Parameter Optimization Deep Learning Retrieval Systems Associative Memory

DeepSeek Publishes mHC to Stabilize Wider Residual Streams in LLMs

The arXiv paper details a Birkhoff‑polytope constraint that kept residual signals stable with modest training overhead.

WhoFi System Re-Identifies People With 95.5% Accuracy by Sensing Wi-Fi Distortions