Multitask reinforcement learning on the distribution of MDPs
By Fumihide Tanaka, Masayuki Yamamura. 2003
76 citations recorded by the source corpus.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Matthew E. Taylor
- Peter Stone
- Joelle Pineau
- Sergey Levine
- Amy Zhang
- Pieter Abbeel
- Shimon Whiteson
- Shagun Sodhani
In Computer Vision and Pattern Recognition
- Amy Zhang
- Joelle Pineau
- Peter Stone
- Shagun Sodhani
- Matthew E. Taylor
- Mykel J. Kochenderfer
- Jayesh K. Gupta
- Bohan Wu
Citations in the corpus
- Model Primitive Hierarchical Lifelong Reinforcement Learning (2019, background): “This setting is different from multi-task reinforcement learning [25, 27, 31] and various meta-reinforcement learning settings [7, 8], where the agent jointly trains on multiple task environments.”
- Sparse multi-task reinforcement learning (2014, unknown): “In [23] different shaping function approaches for Q-table initialization are considered and empirically evaluated, while a model-based approach that estimates statistical information on the distribution of theQ-values is proposed in [25].”
- Multi-Task Reinforcement Learning with Context-based Representations (2021, background): “Multi-task reinforcement learning (MTRL) is a promising approach to train effective real-world agents (Tanaka & Yamamura, 2003; Rusu et al., 2016; Borsa et al., 2016; Rajeswaran et al., 2016; El Bsat et al., 2017; Andreas et al., 2017; Igl et al., 2020; D’Eramo et al., 2020; Yu et al., 2020a).”
- Transfer Learning for Reinforcement Learning Domains: A Survey (2009, background): “Of the thirty-four transfer methods discussed, only five (Tanaka and Yamamura, 2003; Sunmola and Wyatt, 2006; Ferguson and Mahadevan, 2006; Lazaric, 2008; Wilson et al., 2007) attempt to discover internal learning parameters (e.”
- Autonomous Agents and Multi-agent Systems Manuscript No. Learning Potential Functions and Their Representations for Multi-task Reinforcement Learning (unknown): “In [74], the average optimal Q-function of previously experienced tasks is used to initialize the model of a new task.”