Zobrazeno 1 - 10
of 58
pro vyhledávání: '"Ostrovski, Georg"'
Autor:
Nikishin, Evgenii, Oh, Junhyuk, Ostrovski, Georg, Lyle, Clare, Pascanu, Razvan, Dabney, Will, Barreto, André
A growing body of evidence suggests that neural networks employed in deep reinforcement learning (RL) gradually lose their plasticity, the ability to learn from new data; however, the analysis and mitigation of this phenomenon is hampered by the comp
Externí odkaz:
http://arxiv.org/abs/2305.15555
Autor:
Rowland, Mark, Munos, Rémi, Azar, Mohammad Gheshlaghi, Tang, Yunhao, Ostrovski, Georg, Harutyunyan, Anna, Tuyls, Karl, Bellemare, Marc G., Dabney, Will
We analyse quantile temporal-difference learning (QTD), a distributional reinforcement learning algorithm that has proven to be a key component in several successful large-scale applications of reinforcement learning. Despite these empirical successe
Externí odkaz:
http://arxiv.org/abs/2301.04462
Autor:
Gulcehre, Caglar, Srinivasan, Srivatsan, Sygnowski, Jakub, Ostrovski, Georg, Farajtabar, Mehrdad, Hoffman, Matt, Pascanu, Razvan, Doucet, Arnaud
Deep neural networks are the most commonly used function approximators in offline reinforcement learning. Prior works have shown that neural nets trained with TD-learning and gradient descent can exhibit implicit regularization that can be characteri
Externí odkaz:
http://arxiv.org/abs/2207.02099
We identify and study the phenomenon of policy churn, that is, the rapid change of the greedy policy in value-based reinforcement learning. Policy churn operates at a surprisingly rapid pace, changing the greedy action in a large fraction of states w
Externí odkaz:
http://arxiv.org/abs/2206.00730
Learning to act from observational data without active environmental interaction is a well-known challenge in Reinforcement Learning (RL). Recent approaches involve constraints on the learned policy or conservative updates, preventing strong deviatio
Externí odkaz:
http://arxiv.org/abs/2110.14020
Exploration remains a central challenge for reinforcement learning (RL). Virtually all existing methods share the feature of a monolithic behaviour policy that changes only gradually (at best). In contrast, the exploratory behaviours of animals and h
Externí odkaz:
http://arxiv.org/abs/2108.11811
Scaling issues are mundane yet irritating for practitioners of reinforcement learning. Error scales vary across domains, tasks, and stages of learning; sometimes by many orders of magnitude. This can be detrimental to learning speed and stability, cr
Externí odkaz:
http://arxiv.org/abs/2105.05347
While auxiliary tasks play a key role in shaping the representations learnt by reinforcement learning agents, much is still unknown about the mechanisms through which this is achieved. This work develops our understanding of the relationship between
Externí odkaz:
http://arxiv.org/abs/2102.13089
Recent work on exploration in reinforcement learning (RL) has led to a series of increasingly complex solutions to the problem. This increase in complexity often comes at the expense of generality. Recent empirical studies suggest that, when applied
Externí odkaz:
http://arxiv.org/abs/2006.01782
Autor:
Schaul, Tom, Borsa, Diana, Ding, David, Szepesvari, David, Ostrovski, Georg, Dabney, Will, Osindero, Simon
Determining what experience to generate to best facilitate learning (i.e. exploration) is one of the distinguishing features and open challenges in reinforcement learning. The advent of distributed agents that interact with parallel instances of the
Externí odkaz:
http://arxiv.org/abs/1912.06910