COOM: a game benchmark for continual reinforcement learning
The advancement of continual reinforcement learning (RL) has been facing various
obstacles, including standardized metrics and evaluation protocols, demanding …
obstacles, including standardized metrics and evaluation protocols, demanding …
Self-supervised attention-aware reinforcement learning
Visual saliency has emerged as a major visualization tool for interpreting deep
reinforcement learning (RL) agents. However, much of the existing research uses it as an …
reinforcement learning (RL) agents. However, much of the existing research uses it as an …
Intrinsically Motivated and Interactive Reinforcement Learning: a Developmental Approach
P Fournier - 2019 - theses.hal.science
Reinforcement learning (RL) is today more popular than ever, but certain basic skills are still
out of reach of this paradigm: object manipulation, sensorimotor control, natural interaction …
out of reach of this paradigm: object manipulation, sensorimotor control, natural interaction …
[图书][B] Bridging State and Action: Towards Continual Reinforcement Learning
K Khetarpal - 2022 - search.proquest.com
The goal of this thesis is to improve the capability of AI agents to efficiently represent
knowledge and use it to plan and adapt to changes in their environment, through learning …
knowledge and use it to plan and adapt to changes in their environment, through learning …
Multi-task Hierarchical Reinforcement Learning for Compositional Tasks
S Sohn - 2021 - deepblue.lib.umich.edu
This thesis presents the algorithms for solve multiple compositional tasks with high sample
efficiency and strong generalization ability. Central to this work is the subtask graph which …
efficiency and strong generalization ability. Central to this work is the subtask graph which …
Interpretable Continual Learning
We present a framework for interpretable continual learning (ICL). We show that
explanations of previously performed tasks can be used to improve performance on future …
explanations of previously performed tasks can be used to improve performance on future …
Learning generalized temporal abstractions across both action and perception
K Khetarpal - Proceedings of the AAAI Conference on Artificial …, 2019 - aaai.org
Learning temporal abstractions which are partial solutions to a task and could be reused for
other similar or even more complicated tasks is intuitively an ingredient which can help …
other similar or even more complicated tasks is intuitively an ingredient which can help …