Clipping in neurocontrol by adaptive dynamic programming
- Submitting institution
-
City, University of London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 737
- Type
- D - Journal article
- DOI
-
10.1109/TNNLS.2014.2297991
- Title of journal
- IEEE Transactions on Neural Networks and Learning Systems
- Article number
- -
- First page
- 1909
- Volume
- 25
- Issue
- 10
- ISSN
- 2162-237X
- Open access status
- Out of scope for open access requirements
- Month of publication
- January
- Year of publication
- 2014
- URL
-
-
- 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
- 5
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Output formulates a mechanism that reaches the optimum on long-distance predictions when minimising a total cost function by truncating (clipping) the final time step of a trajectory. In so doing, we solve a problem that has limited the application of adaptive dynamic programming and reinforcement learning algorithms to control problems in discrete environments. The first version of the paper was spotlighted by the IEEE Computational Intelligence Society as one of the two best papers in Neural Networks and Learning Systems of 2013.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
