Zobrazeno 1 - 9
of 9
pro vyhledávání: '"Yuu Jinnai"'
Publikováno v:
AAAI
State abstraction can give rise to models of environments that are both compressed and useful, thereby enabling efficient sequential decision making. In this work, we offer the first formalism and analysis of the trade-off between compression and per
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3b1da1906ab6b8d042a45d07c90dad08
https://doi.org/10.1609/aaai.v33i01.33013134
https://doi.org/10.1609/aaai.v33i01.33013134
Autor:
Erwan Lecarpentier, David Abel, Kavosh Asadi, Yuu Jinnai, Emmanuel Rachelson, Michael L. Littman
We consider the problem of knowledge transfer when an agent is facing a series of Reinforcement Learning (RL) tasks. We introduce a novel metric between Markov Decision Processes (MDPs) and establish that close MDPs have close optimal value functions
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e9b9b14c297dbb8e6c19c02799cd0b94
Autor:
Yuu Jinnai, Alex Fukunaga
Publikováno v:
Journal of Artificial Intelligence Research. 60:491-548
Parallel best-first search algorithms such as Hash Distributed A* (HDA*) distribute work among the processes using a global hash function. We analyze the search and communication overheads of state-of-the-art hash-based parallel best-first search alg
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bd0228038296f24393bf308d00f9df60
https://doi.org/10.1613/jair.5225
https://doi.org/10.1613/jair.5225
Autor:
Yuu Jinnai, Alex Fukunaga
Publikováno v:
Proceedings of the International Conference on Automated Planning and Scheduling. 26:184-192
Hash Distributed A* (HDA*) is an efficient parallel best first algorithm that asynchronously distributes work among the processes using a global hash function. We investigate domain-independent methods for automatically creating effective work distri
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::93197f5812169d434c1de2faa91b23eb
https://doi.org/10.1609/icaps.v26i1.13747
https://doi.org/10.1609/icaps.v26i1.13747
Publikováno v:
AAAI
Neural Architecture Search (NAS) has shown great success in automating the design of neural networks, but the prohibitive amount of computations behind current NAS methods requires further investigations in improving the sample efficiency and the net
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::79753380528bb9f3ee1225dc19d5848b
Publikováno v:
Handbook of Parallel Constraint Reasoning ISBN: 9783319635156
Handbook of Parallel Constraint Reasoning
Handbook of Parallel Constraint Reasoning
A* is a best-first search algorithm for finding optimal-cost paths in graphs. A* benefits significantly from parallelism because in many applications, A* is limited by memory usage, so distributed memory implementations of A* that use all of the aggr
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::5425e127dac96ea6edd36be365011e38
https://doi.org/10.1007/978-3-319-63516-3_11
https://doi.org/10.1007/978-3-319-63516-3_11
Autor:
Yuu Jinnai, Alex Fukunaga
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 31
Black-box domains where the successor states generated by applying an action are generated by a completely opaque simulator pose a challenge for domain-independent planning. The main computational bottleneck in search-based planning for such domains
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::441abdad129a33e76c423a5f189ad992
https://doi.org/10.1609/aaai.v31i1.11118
https://doi.org/10.1609/aaai.v31i1.11118
Autor:
Yuu Jinnai, Alex Fukunaga
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 31
Black-box domains where the successor states generated by applying an action are generated by a completely opaque simulator pose a challenge for domain-independent planning. The main computational bottleneck in search-based planning for such domains
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::534e246936b9ef0b0c147ca8dd4beaaa
https://doi.org/10.1609/aaai.v31i1.10663
https://doi.org/10.1609/aaai.v31i1.10663
Autor:
Yuu Jinnai, Alex Fukunaga
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 30
Hash Distributed A* (HDA*) is an efficient parallel best first algorithm that asynchronously distributes work among the processes using a global hash function. Although Zobrist hashing, the standard hash function used by HDA*, achieves good load bala
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::b7b5d1f0a78556aa4db895d2068961fc
https://doi.org/10.1609/aaai.v30i1.10065
https://doi.org/10.1609/aaai.v30i1.10065