Chinese content研究揭示随机环境下强化学习可塑性损失的影响机制
这篇论文把 SAC 和 DeepMind Control 任务加上噪声做实验，讲清了什么时候该用可塑性保持方法、什么时候重置就够了，做 RL 的可以看看。
#DeepMind Control
3 stories · 1 in 7 days · 1 in 30 days
Stories, products and papers tagged "DeepMind Control", newest first.
Nov 2
Sep 28
StarWM Routes Attention for Robust World Model Training
The technique targets a real failure mode: reconstruction-based world models waste capacity on pixel area instead of dynamics relevance, so agents get distracted by background video. Builders training RL agents or world models can borrow the attention-routed reconstruction pattern to keep representations focused on task-relevant content. It is a paper-level method benchmarked on DeepMind Control, comparable against both reconstruction-based and reconstruction-free baselines.
Jul 14
Papers09:21
Chinese content深度何时能在组合中存活？潜在世界模型中的计算-质量状态这篇论文用实验告诉你，深的世界模型在滚动预测时不一定管用——在DeepMind Control的9个任务里，有2个反而是早退浅层更好，而且原因竟然出在训练方式上。搞模型规划的值得看看。