Results 131 to 140 of about 685,298 (326)
Advanced Materials, EarlyView.The SciAgents AI model drives hypothesis generation by harnessing multi‐agent graph reasoning, extracting insights from knowledge graphs constructed from scientific papers. Each agent plays a specific role: the Ontologist defines concepts, the Scientists draft and refine proposals, and the Critic reviews.
Alireza Ghafarollahi, Markus J. Buehler
wiley +1 more source
Advanced Materials, EarlyView.Protein can undergo liquid–liquid phase separation and liquid‐to‐solid transition to form liquid condensates and solid aggregates. These phase transitions can be influenced by post‐translational modifications, mutations, and various environmental factors.
Tianchen Li +3 more
wiley +1 more source
A Survey Analyzing Generalization in Deep Reinforcement Learning [PDF]
arXivReinforcement learning research obtained significant success and attention with the utilization of deep neural networks to solve problems in high dimensional state or action spaces. While deep reinforcement learning policies are currently being deployed in many different fields from medical applications to large language models, there are still ongoing
arxiv
A View on Deep Reinforcement Learning in System Optimization [PDF]
arXiv, 2019 Many real-world systems problems require reasoning about the long term consequences of actions taken to configure and manage the system. These problems with delayed and often sequentially aggregated reward, are often inherently reinforcement learning problems and present the opportunity to leverage the recent substantial advances in deep reinforcement ...
arxiv
Engineering CAR‐T Therapeutics for Enhanced Solid Tumor Targeting
Advanced Materials, EarlyView.CART cell therapy has proven effective for blood cancers but struggles with solid tumors due to diverse antigens and complex environments. Recent efforts focus on improving CAR design and validation platforms. Advances in protein engineering, machine learning, and organoid systems aim to enhance CAR‐T therapy against solid tumors.
Danqing Zhu +4 more
wiley +1 more source
Deep Reinforcement Learning and the Deadly Triad
, 2018 We know from reinforcement learning theory that temporal difference learning can fail in certain cases. Sutton and Barto (2018) identify a deadly triad of function approximation, bootstrapping, and off-policy learning. When these three properties are combined, learning can diverge with the value estimates becoming unbounded. However, several algorithms
van Hasselt, Hado +5 more
openaire +4 more sources
Deep Reinforcement and InfoMax Learning
, 2020 NeurIPS ...
Mazoure, Bogdan +4 more
openaire +2 more sources
Advanced Materials, EarlyView.This article investigates the micromechanics of bamboo epidermis, focusing on how anisotropic silica particle distributions enhance toughness. By integrating experimental imaging, 3D printing, and generative AI, the study develops bio‐inspired particle‐reinforced composites with mechanical properties akin to bamboo.
Zhao Qin, Aymeric Pierre Destree
wiley +1 more source
Deep reinforcement learning from human preferences
, 2017 For sophisticated reinforcement learning (RL) systems to interact usefully with real-world environments, we need to communicate complex goals to these systems.
Amodei, Dario +5 more
core
Counterexample-Guided Repair of Reinforcement Learning Systems Using Safety Critics [PDF]
arXivNaively trained Deep Reinforcement Learning agents may fail to satisfy vital safety constraints. To avoid costly retraining, we may desire to repair a previously trained reinforcement learning agent to obviate unsafe behaviour. We devise a counterexample-guided repair algorithm for repairing reinforcement learning systems leveraging safety critics. The
arxiv