Zobrazeno 1 - 10
of 37
pro vyhledávání: '"Flennerhag, Sebastian"'
Autor:
Baumli, Kate, Baveja, Satinder, Behbahani, Feryal, Chan, Harris, Comanici, Gheorghe, Flennerhag, Sebastian, Gazeau, Maxime, Holsheimer, Kristian, Horgan, Dan, Laskin, Michael, Lyle, Clare, Masoom, Hussain, McKinney, Kay, Mnih, Volodymyr, Neitz, Alexander, Nikulin, Dmitry, Pardo, Fabio, Parker-Holder, Jack, Quan, John, Rocktäschel, Tim, Sahni, Himanshu, Schaul, Tom, Schroecker, Yannick, Spencer, Stephen, Steigerwald, Richie, Wang, Luyu, Zhang, Lei
Building generalist agents that can accomplish many goals in rich open-ended environments is one of the research frontiers for reinforcement learning. A key limiting factor for building generalist agents with RL has been the need for a large number o
Externí odkaz:
http://arxiv.org/abs/2312.09187
We work towards a unifying paradigm for accelerating policy optimization methods in reinforcement learning (RL) by integrating foresight in the policy improvement step via optimistic and adaptive updates. Leveraging the connection between policy iter
Externí odkaz:
http://arxiv.org/abs/2306.10587
Autor:
Lange, Robert Tjarko, Schaul, Tom, Chen, Yutian, Lu, Chris, Zahavy, Tom, Dalibard, Valentin, Flennerhag, Sebastian
Genetic algorithms constitute a family of black-box optimization algorithms, which take inspiration from the principles of biological evolution. While they provide a general-purpose tool for optimization, their particular instantiations can be heuris
Externí odkaz:
http://arxiv.org/abs/2304.03995
Autor:
Moskovitz, Ted, O'Donoghue, Brendan, Veeriah, Vivek, Flennerhag, Sebastian, Singh, Satinder, Zahavy, Tom
In recent years, Reinforcement Learning (RL) has been applied to real-world problems with increasing success. Such applications often require to put constraints on the agent's behavior. Existing algorithms for constrained RL (CRL) rely on gradient de
Externí odkaz:
http://arxiv.org/abs/2302.01275
Autor:
Flennerhag, Sebastian, Zahavy, Tom, O'Donoghue, Brendan, van Hasselt, Hado, György, András, Singh, Satinder
We study the connection between gradient-based meta-learning and convex op-timisation. We observe that gradient descent with momentum is a special case of meta-gradients, and building on recent results in optimisation, we prove convergence rates for
Externí odkaz:
http://arxiv.org/abs/2301.03236
Autor:
Lange, Robert Tjarko, Schaul, Tom, Chen, Yutian, Zahavy, Tom, Dallibard, Valentin, Lu, Chris, Singh, Satinder, Flennerhag, Sebastian
Publikováno v:
11th International Conference on Learning Representations, ICLR 2023
Optimizing functions without access to gradients is the remit of black-box methods such as evolution strategies. While highly general, their learning dynamics are often times heuristic and inflexible - exactly the limitations that meta-learning can a
Externí odkaz:
http://arxiv.org/abs/2211.11260
We evaluate the use of original game curricula supported by the Atari 2600 console as a heterogeneous transfer benchmark for deep reinforcement learning agents. Game designers created curricula using combinations of several discrete modifications to
Externí odkaz:
http://arxiv.org/abs/2210.12448
Autor:
Luketina, Jelena, Flennerhag, Sebastian, Schroecker, Yannick, Abel, David, Zahavy, Tom, Singh, Satinder
Meta-gradient methods (Xu et al., 2018; Zahavy et al., 2020) offer a promising solution to the problem of hyperparameter selection and adaptation in non-stationary reinforcement learning problems. However, the properties of meta-gradients in such env
Externí odkaz:
http://arxiv.org/abs/2209.06159
Autor:
Zahavy, Tom, Schroecker, Yannick, Behbahani, Feryal, Baumli, Kate, Flennerhag, Sebastian, Hou, Shaobo, Singh, Satinder
Finding different solutions to the same problem is a key aspect of intelligence associated with creativity and adaptation to novel situations. In reinforcement learning, a set of diverse policies can be useful for exploration, transfer, hierarchy, an
Externí odkaz:
http://arxiv.org/abs/2205.13521
Autor:
Kirsch, Louis, Flennerhag, Sebastian, van Hasselt, Hado, Friesen, Abram, Oh, Junhyuk, Chen, Yutian
Meta reinforcement learning (RL) attempts to discover new RL algorithms automatically from environment interaction. In so-called black-box approaches, the policy and the learning algorithm are jointly represented by a single neural network. These met
Externí odkaz:
http://arxiv.org/abs/2109.10781