02/02/2021

Encoding Human Domain Knowledge to Warm Start Reinforcement Learning

Andrew Silva, Matthew Gombolay

Keywords:

Abstract Paper Similar Papers

Abstract: Deep reinforcement learning has been successful in a variety of tasks, such as game playing and robotic manipulation. However, attempting to learn tabula rasa disregards the logical structure of many domains as well as the wealth of readily available knowledge from domain experts that could help "warm start" the learning process. We present a novel reinforcement learning technique that allows for intelligent initialization of a neural network weights and architecture. Our approach permits the encoding domain knowledge directly into a neural decision tree, and improves upon that knowledge with policy gradient updates. We empirically validate our approach on two OpenAI Gym tasks and two modified StarCraft 2 tasks, showing that our novel architecture outperforms multilayer-perceptron and recurrent architectures. Our knowledge-based framework finds superior policies compared to imitation learning-based and prior knowledge-based approaches. Importantly, we demonstrate that our approach can be used by untrained humans to initially provide >80% increase in expected reward relative to baselines prior to training (p < 0.001), which results in a >60% increase in expected reward after policy optimization (p = 0.011).

The video of this talk cannot be embedded. You can watch it here:
https://slideslive.com/38948570
(Link will open in new window)
 0
 0
 0
 0
  Share    
This is an embedded video. Talk and the respective paper are published at AAAI 2021 virtual conference. If you are one of the authors of the paper and want to manage your upload, see the question "My papertalk has been externally embedded..." in the FAQ section.

Comments

Post Comment
no comments yet
code of conduct: tbd Characters remaining: 140

Similar Papers

 0
 0
 0
 0
 3:12 
 0
 0
 0
 0
 14:37 
 0
 0
 0
 0
 5:25 
 0
 0
 0
 0
 3:24 
 0
 0
 0
 0
 1:01 
 0
 0
 0
 0
 1:01 
 0
 0
 0
 0
 3:18 
 0
 0
 0
 1
 10:04 
 0
 0
 0
 0
 5:00 
 0
 0
 0
 0
 8:45 
 0
 0
 0
 0
 12:26 
 0
 0
 0
 0
 4:58 
 0
 0
 0
 0
 5:18 
 0
 0
 0
 0
 3:30 
 0
 0
 0
 0
 5:22 
 0
 0
 0
 0
 3:23 
 0
 0
 0
 0
 3:22 
 0
 0
 0
 0
 10:28 
 0
 0
 0
 0
 10:58 
 0
 0
 0
 0
 1:01 
 0
 0
 0
 0
 4:38 
 0
 0
 0
 0
 11:27 
 0
 0
 0
 0
 3:11 
 0
 0
 0
 0
 3:01 
 0
 0
 0
 0
 10:18 
 0
 0
 0
 0
 3:21 
 0
 0
 0
 0
 5:35 
 0
 0
 0
 1
 4:49 
 0
 0
 0
 0
 15:04 
 0
 0
 0
 0
 18:00 
 0
 0
 0
 0
 4:14 
 0
 0
 0
 0
 3:58 
 0
 0
 0
 0
 5:01 
 0
 0
 0
 0
 2:46 
 0
 0
 0
 0
 5:15 
 0
 0
 0
 0
 5:17 
 0
 0
 0
 0
 5:11 
 0
 0
 0
 0
 4:34 
 0
 0
 0
 0
 6:26 
 0
 0
 0
 0
 15:38