Independent learning achieves approximate Nash equilibria in partially observable Markov potential games with decoupled dynamics and near-polynomial complexity via finite history windows.
DiPhon defines graphon diffusion via a Jacobi SDE for scalable graph generation, matching first moments exactly and preserving topology across sizes without retraining.
Dithered randomized Hadamard quantization is unbiased and achieves mean squared error asymptotically matching dense random rotations at O(d log d) cost.
MyoChallenge 2025 benchmarks musculoskeletal sports control via simulated table tennis and soccer tasks, advancing agile motor algorithms across 70 teams.
Standard pass@k scaling laws suffer statistical shortcomings, so a beta-binomial framework and dynamic sampling strategy more accurately predict rare LLM capabilities and risks from limited data.
Neural LoFi frames deep training as iterative spectral low-degree filtering, predicting layer-wise feature selection, concept emergence, and compositional depth via low-degree correlation dynamics.