From AlphaGo to Power System AI: What Engineers Can Learn from Solving the Most Complex Board Game
Authors:
- 19 February 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Power and Energy Magazine
- Vol. 16 (2) , 76-84
Abstract
Since its early development in the 1950s, artificial intelligence (AI) has been studied to mimic the function of human brains in solving science and engineering problems. Despite alternating bust and boom periods, AI has consistently progressed with a number of development milestones such as evolutionary algorithms, neural networks, and deep learning. For many AI test applications, the game of Go has been long considered one of the hardest problems and the most difficult among all major board games for AI to solve or mimic the way that the human brain functions.Keywords
This publication has 6 references indexed in Scilit:
- Risk Assessment of Multi-Timescale Cascading Outages Based on Markovian Tree SearchIEEE Transactions on Power Systems, 2016
- Deep learning neural network for power system fault diagnosis2016
- Mastering the game of Go with deep neural networks and tree searchNature, 2016
- Enabling Reliability-Differentiated Service in Residential Distribution Networks With PHEVs: A Hierarchical Game ApproachIEEE Transactions on Smart Grid, 2015
- Online Static Security Assessment Module Using Artificial Neural NetworksIEEE Transactions on Power Systems, 2013
- All Systems GoThe Sciences, 1998