Timezone: »

 
Poster
Denoised MDPs: Learning World Models Better Than the World Itself
Tongzhou Wang · Simon Du · Antonio Torralba · Phillip Isola · Amy Zhang · Yuandong Tian

Wed Jul 20 03:30 PM -- 05:30 PM (PDT) @ Hall E #829

The ability to separate signal from noise, and reason with clean abstractions, is critical to intelligence. With this ability, humans can efficiently perform real world tasks without considering all possible nuisance factors. How can artificial agents do the same? What kind of information can agents safely discard as noises? In this work, we categorize information out in the wild into four types based on controllability and relation with reward, and formulate useful information as that which is both controllable and reward-relevant. This framework clarifies the kinds information removed by various prior work on representation learning in reinforcement learning (RL), and leads to our proposed approach of learning a Denoised MDP that explicitly factors out certain noise distractors. Extensive experiments on variants of DeepMind Control Suite and RoboDesk demonstrate superior performance of our denoised world model over using raw observations alone, and over prior works, across policy optimization control tasks as well as the non-control task of joint position regression.

Author Information

Tongzhou Wang (MIT)
Simon Du (University of Washington)
Antonio Torralba (MIT)
Phillip Isola (MIT)
Amy Zhang (FAIR / UC Berkeley)
Yuandong Tian (Facebook AI Research)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors