• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Student
Title
Supervisor
Faculty
Educational Programme
Final Grade
Year of Graduation
Dmitriy Pyrkin
Reinforcement Learning for Environments With Sparse Actions
2019
Intrinsic rewards are a key tool to efficient exploration in deep reinforcement learning. However, using such rewards requires one to balance the relative importances of the two rewards.

Overvalued intrinsic rewards can cause agent to abandon the original objective and fail to solve the environment. This instability is caused by agent maximizing a weighted combination of extrinsic and intrinsic rewards.

We propose a new method for combining extrinsic and intrinsic rewards. Our method allows using intrinsic rewards with a limit on how far agent can deviate from the currently learned policy during one training episode.

We demonstrate the effectiveness of our approach in both discrete and continuous environments.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses