Why most decisions are easy in Tetris—And perhaps in other sequential decision problems, as well
- Submitting institution
-
The University of Bath
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 158353776
- Type
- E - Conference contribution
- DOI
-
-
- Title of conference / published proceedings
- Proceedings of The 33rd International Conference on Machine Learning
- First page
- 1757
- Volume
- -
- Issue
- -
- ISSN
- 2640-3498
- Open access status
- Deposit exception
- Month of publication
- June
- Year of publication
- 2016
- URL
-
http://proceedings.mlr.press/v48/simsek16.html
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
2
- Research group(s)
-
-
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This is a significant result that opens up a new area of investigation in reinforcement learning, with potential to drastically improve the efficiency of learning (as measured by the number of observations and the amount of computation needed). In the game of Tetris, the approach reduced the average branching factor from 17 to 1.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -