Carlos Florensa Campo and David Held and Markus Wulfmeier and Michael Zhang and Pieter Abbeel

EECS Department, University of California, Berkeley

Technical Report No. UCB/EECS-2018-162

December 1, 2018

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-162.pdf

Many relevant tasks require an agent to reach a certain state, or to manipulate objects into a desired configuration. For example, we might want a robot to align and assemble a gear onto an axle or insert and turn a key in a lock. These goal-oriented tasks present a considerable challenge for reinforcement learning, since their natural reward function is sparse and prohibitive amounts of exploration are required to reach the goal and receive some learning signal. Past approaches tackle these problems by exploiting expert demonstrations or by manually designing a task-specific reward shaping function to guide the learning agent. Instead, we propose a method to learn these tasks without requiring any prior knowledge other than obtaining a single state in which the task is achieved. The robot is trained in reverse, gradually learning to reach the goal from a set of start states increasingly far from the goal. Our method automatically generates a curriculum of start states that adapts to the agent's performance, leading to efficient training on goal-oriented tasks. We demonstrate our approach on difficult simulated navigation and fine-grained manipulation problems, not solvable by state-of-the-art reinforcement learning methods.

Advisors: Pieter Abbeel

---

BibTeX citation:

@mastersthesis{Florensa Campo:EECS-2018-162,
    Author= {Florensa Campo, Carlos and Held, David and Wulfmeier,  Markus and Zhang, Michael and Abbeel, Pieter},
    Title= {Reverse Curriculum Generation for Reinforcement Learning},
    School= {EECS Department, University of California, Berkeley},
    Year= {2018},
    Month= {Dec},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-162.html},
    Number= {UCB/EECS-2018-162},
    Abstract= {Many relevant tasks require an agent to reach a certain state, or to manipulate objects into a desired configuration. For example, we might want a robot to align and assemble a gear onto an axle or insert and turn a key in a lock. These goal-oriented tasks present a considerable challenge for reinforcement learning, since their natural reward function is sparse and prohibitive amounts of exploration are required to reach the goal and receive some learning signal. Past approaches tackle these problems by exploiting expert demonstrations or by manually designing a task-specific reward shaping function to guide the learning agent. Instead, we propose a method to learn these tasks without requiring any prior knowledge other than obtaining a single state in which the task is achieved. The robot is trained in reverse, gradually learning to reach the goal from a set of start states increasingly far from the goal. Our method automatically generates a curriculum of start states that adapts to the agent's performance, leading to efficient training on goal-oriented tasks. We demonstrate our approach on difficult simulated navigation and fine-grained manipulation problems, not solvable by state-of-the-art reinforcement learning methods.},
}

EndNote citation:

%0 Thesis
%A Florensa Campo, Carlos 
%A Held, David 
%A Wulfmeier,  Markus 
%A Zhang, Michael 
%A Abbeel, Pieter 
%T Reverse Curriculum Generation for Reinforcement Learning
%I EECS Department, University of California, Berkeley
%D 2018
%8 December 1
%@ UCB/EECS-2018-162
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-162.html
%F Florensa Campo:EECS-2018-162