Publications
agents ai data decisions esports gaist games heuristics hyperheuristics optimisation scheduling search
2018 |
Baier, Hendrik; Cowling, Peter I Evolutionary MCTS with Flexible Search Horizon Inproceedings Proceedings of the AAAI Artificial Intelligence and Interactive Digital Entertainment Conference 2018 (AIIDE'18), Canada, AAAI, 2018. Abstract | BibTeX | Tags: agents, ai, decisions, games, heuristics, search | Links: @inproceedings{baier2018evolutionary, title = {Evolutionary MCTS with Flexible Search Horizon}, author = {Hendrik Baier and Peter I. Cowling}, url = {http://www.petercowling.com/wp-content/uploads/2018/09/Evolutionary-MCTS-with-Flexible-Search-Horizon.pdf}, year = {2018}, date = {2018-11-07}, booktitle = {Proceedings of the AAAI Artificial Intelligence and Interactive Digital Entertainment Conference 2018 (AIIDE'18), Canada}, journal = {In Proceedings of Artificial Intelligence and Interactive Digital Entertainment (AIIDE'18), Canada}, publisher = {AAAI}, abstract = {In turn-based multi-action adversarial games each player turn consists of several atomic actions, resulting in an extremely high branching factor. Many strategy board, card, and video games fall into this category, which is currently best played by Evolutionary MCTS (EMCTS) – searching a tree with nodes representing action sequences as genomes, and edges representing mutations of those genomes. However, regular EMCTS is unable to search beyond the current player’s turn, leading to strategic short-sightedness. In this paper, we extend EMCTS to search to any given search depth beyond the current turn, using simple models of its own and the opponent’s behavior. Experiments on the game Hero Academy show that this Flexible-Horizon EMCTS (FH-EMCTS) convincingly out- performs several baselines including regular EMCTS, Online Evolutionary Planning (OEP), and vanilla MCTS, at all tested numbers of atomic actions per turn. Additionally, the separate contributions of the behavior models and the flexible search horizon are analyzed.}, keywords = {agents, ai, decisions, games, heuristics, search}, pubstate = {published}, tppubtype = {inproceedings} } In turn-based multi-action adversarial games each player turn consists of several atomic actions, resulting in an extremely high branching factor. Many strategy board, card, and video games fall into this category, which is currently best played by Evolutionary MCTS (EMCTS) – searching a tree with nodes representing action sequences as genomes, and edges representing mutations of those genomes. However, regular EMCTS is unable to search beyond the current player’s turn, leading to strategic short-sightedness. In this paper, we extend EMCTS to search to any given search depth beyond the current turn, using simple models of its own and the opponent’s behavior. Experiments on the game Hero Academy show that this Flexible-Horizon EMCTS (FH-EMCTS) convincingly out- performs several baselines including regular EMCTS, Online Evolutionary Planning (OEP), and vanilla MCTS, at all tested numbers of atomic actions per turn. Additionally, the separate contributions of the behavior models and the flexible search horizon are analyzed. |
Block, Florian; Hodge, Victoria; Hobson, Stephen; Sephton, Nick; Devlin, Sam; Ursu, Marian F; Drachen, Anders; Cowling, Peter I Narrative Bytes: Data-Driven Content Production in Esports Inproceedings Proceedings of the 2018 ACM International Conference on Interactive Experiences for TV and Online Video, pp. 29–41, ACM 2018. Abstract | BibTeX | Tags: data, esports, games | Links: @inproceedings{block2018narrative, title = {Narrative Bytes: Data-Driven Content Production in Esports}, author = {Florian Block and Victoria Hodge and Stephen Hobson and Nick Sephton and Sam Devlin and Marian F Ursu and Anders Drachen and Peter I Cowling}, url = {http://www.petercowling.com/wp-content/uploads/2018/08/block2018narrative.pdf}, doi = {10.1145/3210825.3210833}, year = {2018}, date = {2018-01-01}, booktitle = {Proceedings of the 2018 ACM International Conference on Interactive Experiences for TV and Online Video}, pages = {29--41}, organization = {ACM}, abstract = {Esports – video games played competitively that are broadcast to large audiences – are a rapidly growing new form of mainstream entertainment. Esports borrow from traditional TV, but are a qualitatively different genre, due to the high flexibility of content capture and availability of detailed gameplay data. Indeed, in esports, there is access to both real-time and historical data about any action taken in the virtual world. This aspect motivates the research presented here, the question asked being: can the information buried deep in such data, unavailable to the human eye, be unlocked and used to improve the live broadcast compilations of the events? In this paper, we present a largescale case study of a production tool called Echo, which we developed in close collaboration with leading industry stakeholders. Echo uses live and historic match data to detect extraordinary player performances in the popular esport Dota 2, and dynamically translates interesting data points into audience-facing graphics. Echo was deployed at one of the largest yearly Dota 2 tournaments, which was watched by 25 million people. An analysis of 40 hours of video, over 46,000 live chat messages, and feedback of 98 audience members showed that Echo measurably affected the range and quality of storytelling, increased audience engagement, and invoked rich emotional response among viewers. }, keywords = {data, esports, games}, pubstate = {published}, tppubtype = {inproceedings} } Esports – video games played competitively that are broadcast to large audiences – are a rapidly growing new form of mainstream entertainment. Esports borrow from traditional TV, but are a qualitatively different genre, due to the high flexibility of content capture and availability of detailed gameplay data. Indeed, in esports, there is access to both real-time and historical data about any action taken in the virtual world. This aspect motivates the research presented here, the question asked being: can the information buried deep in such data, unavailable to the human eye, be unlocked and used to improve the live broadcast compilations of the events? In this paper, we present a largescale case study of a production tool called Echo, which we developed in close collaboration with leading industry stakeholders. Echo uses live and historic match data to detect extraordinary player performances in the popular esport Dota 2, and dynamically translates interesting data points into audience-facing graphics. Echo was deployed at one of the largest yearly Dota 2 tournaments, which was watched by 25 million people. An analysis of 40 hours of video, over 46,000 live chat messages, and feedback of 98 audience members showed that Echo measurably affected the range and quality of storytelling, increased audience engagement, and invoked rich emotional response among viewers. |