Theory

Learning theory, optimisation theory, generalisation, mathematical analysis of models

8 articles

All Agents Language models Reinforcement learning Vision Generative media Speech & audio Robotics & embodied AI Safety & security Evaluation & benchmarks Efficiency & systems Code & software AI for science & health Theory

Viewgrid list compact

Polylogarithmic Nash regret achieved for any finite matrix game

Finite probe sets identify neural networks, but hidden activations make them far more informative

Unmodified posterior sampling achieves minimax regret in reinforcement learning

Muon Splits Loss Stability From Update Reversal During LLM Pretraining

Cyclical monotonicity enforces efficient adversarial transport in robust learning

LLMs generate generalized plans with machine-checked completeness proofs

Frozen agent states can retain causal mechanisms beyond task performance

Shared output errors can saturate networks trained with random feedback