arXiv Machine Learning
Action-Driven Processes for Continuous-Time Control
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Aug 21
Spike-based Belief Propagation in Nonlinear Dynamical Systems
arXiv:2608. 19907v1 Announce Type: new Abstract: This paper presents a Bayesian control framework that integrates spike-based dynamics with probabilistic inference for adaptive control.
Hugging Face Trending Papers
Aug 20
Spike-based Belief Propagation in Nonlinear Dynamical Systems
The paper introduces a Bayesian control framework that merges spike-based neural dynamics with probabilistic inference for adaptive control. It applies this brain-inspired model to the mountain car parking problem, showing that the controller can update states in real time and generate goal-directed action plans via spike-driven dynamics. The results suggest the model could serve as a bridge between computational neuroscience and probabilistic control theory.
arXiv AI
Jun 4
From Ticks to Flows: Dynamics of Neural Reinforcement Learning in Continuous Environments
arXiv:2606. 04275v1 Announce Type: cross Abstract: We present a novel theoretical framework for deep reinforcement learning (RL) in continuous environments by modeling the problem as a continuous-time stochastic process, drawing on insights from stochastic control.
arXiv AI
Oct 1
Diffusion-Augmented Markov Decision Processes for Maximum Entropy Reinforcement Learning
The paper introduces Diffusion-Augmented Markov Decision Processes (DA‑MDPs), a framework that extends Maximum Entropy Reinforcement Learning to diffusion-based policies. DA‑MDPs treat each reverse‑diffusion step as an RL decision, deriving a tractable reverse‑KL bound that decomposes across denoising transitions and yields diffusion‑augmented soft rewards, value functions, and policy objectives. The authors implement this framework with PPO, REPPO, and a maximum‑entropy WPO variant, showing improved continuous‑control performance, higher success rates on manipulation tasks, and memory‑efficient training with action chunking.
arXiv AI
Aug 24
Behavior-Consistent Deep Reinforcement Learning
The paper introduces the concept of behavior-consistent deep reinforcement learning, aiming to produce high-performing policies that remain distributionally similar across different training runs. It shows that maximum-entropy RL can control behavioral divergence by anchoring runs to a common prior, and proves that for Boltzmann policies, a temperature proportional to Q‑function disagreement limits pairwise KL divergence. Building on this, the authors propose Q‑value Expectile Disagreement (QED), a state‑dependent temperature schedule that uses double‑critic disagreement to approximate cross‑run disagreement, and demonstrate that QED reduces across‑run divergence by two orders of magnitude on 18 continuous‑control tasks without sacrificing performance.
arXiv Machine Learning
Aug 6
Distributional Active Inference
arXiv:2601. 20985v2 Announce Type: replace Abstract: Optimal control of complex environments with robotic systems faces two complementary and intertwined challenges: efficient organization of sensory state information and far-sighted action planning.