DeepMind 科技百花 2023-02-16 08:45:26 Graph schemas as abstractions for transfer learning, inference, and planning DownloadView publicationAbstractWe propose schemas as a model for abstractions that can be used fo... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind 大话互联网 2023-02-13 08:24:32 Universal Agent Mixtures and the Geometry of Intelligence DownloadView publicationAbstractInspired by recent progress in multi-agent Reinforcement Learn... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind 大话互联网 2023-02-09 15:21:12 Equivariant MuZero DownloadView publicationAbstractDeep reinforcement learning repeatedly succeeds in closed, wel... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind nightclub 2023-02-09 09:22:58 Scaling Goal-based Exploration via Pruning Proto-goals DownloadView publicationAbstractOne of the gnarliest challenges in reinforcement learning is e... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind 大话互联网 2023-02-07 13:13:10 Exploration via Epistemic Value Estimation DownloadView publicationAbstractHow to efficiently explore in reinforcement learning is an ope... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind 大话互联网 2023-02-07 11:40:44 3D Neural Embedding Likelihood for Robust Sim-to-Real Transfer in Inverse Graphics View publicationAbstractA central challenge in 3D scene perception via inverse graphics is robustly ... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind nightclub 2023-02-03 12:32:41 Diversity Through Exclusion (DTE): Niche Identification for Reinforcement Learning through Value-Decomposition DownloadView publicationAbstractMany environments contain numerous available niches of var... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind 大话互联网 2023-02-02 15:49:34 PGMax: Factor Graphs for Discrete Probabilistic Graphical Models and Loopy Belief Propagation in JAX DownloadView publicationAbstractPGMax is an open-source Python package for easy specification ... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind 大话互联网 2023-02-02 13:25:45 Dual Algorithmic Reasoning DownloadView publicationAbstractNeural Algorithmic Reasoning is an emerging ar... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind nightclub 2023-02-02 10:38:32 Reinforcement Learning for Minimizing Age of Information over Wireless Links View publicationAbstractIn this chapter, we study the Ag... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind 大话互联网 2023-02-02 09:01:46 Learning Noisy OR Bayesian Networks with Max-Product Belief Propagation DownloadView publicationAbstractNoisy-OR Bayesian Networks (BNs) are a family of probabilistic... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
DeepMind 大话互联网 2023-01-29 13:45:41 Distilling Internet-Scale Vision-Language Models into Embodied Agents DownloadView publicationAbstractInstruction-following agents must ground language into their obser... 评论 分享 微信扫一扫：分享 微信点“扫一扫” 点右上“...”，便可分享本文
