Assisted Feature Engineering and Feature Learning to Build Knowledge-Based Agents for Arcade Games. Andelefski, B. & Schiffer, S. In van den Herik, J., Rocha, A. P., & Filipe, J., editors, ICAART 2017 - Proceedings of the 4th International Conference on Agents and Artificial Intelligence, volume 2 - Artificial Intelligence, pages 228–238, February 6-8, 2017. SciTePress. abstract bibtex Human knowledge can greatly increase the performance of autonomous agents. Leveraging this knowledge is sometimes neither straightforward nor easy. In this paper, we present an approach for assisted feature engineering and feature learning to build knowledge-based agents for three arcade games within the Arcade Learning Environment. While existing approaches mostly use model-free approaches we aim at creating a descriptive set of features for world modelling and building agents. To this end, we provide (visual) assistance in identifying and modelling features from RAM, we allow for learning features based on labeled game data, and we allow for creating basic agents using the above features. In our evaluation, we compare different methods to learn features from the RAM. We then compare several agents using different sets of manual and learned features with one another and with the state-of-the-art.
@inproceedings{ Andelefski:Schiffer:ICAART2017:AssistedFeatureLearning,
author = {Bastian Andelefski and Stefan Schiffer},
title = {Assisted Feature Engineering and Feature Learning to Build Knowledge-Based Agents for Arcade Games},
year = {2017},
pages = {228--238},
editor = {Jaap van den Herik and Ana Paula Rocha and Joaquim Filipe},
booktitle = {ICAART 2017 - Proceedings of the 4th International Conference on Agents and Artificial Intelligence},
volume = {2 - Artificial Intelligence},
location = {Porto, Portugal},
month = {February 6-8},
publisher = {SciTePress},
isbn = {978-989-758-220-2},
abstract = {Human knowledge can greatly increase the performance
of autonomous agents. Leveraging this knowledge is
sometimes neither straightforward nor easy. In this
paper, we present an approach for assisted feature
engineering and feature learning to build
knowledge-based agents for three arcade games within
the Arcade Learning Environment. While existing
approaches mostly use model-free approaches we aim
at creating a descriptive set of features for world
modelling and building agents. To this end, we
provide (visual) assistance in identifying and
modelling features from RAM, we allow for learning
features based on labeled game data, and we allow
for creating basic agents using the above
features. In our evaluation, we compare different
methods to learn features from the RAM. We then
compare several agents using different sets of
manual and learned features with one another and
with the state-of-the-art.},
}
Downloads: 0
{"_id":"kJ3fxKgqzCkRtLhTH","bibbaseid":"andelefski-schiffer-assistedfeatureengineeringandfeaturelearningtobuildknowledgebasedagentsforarcadegames-2017","downloads":0,"creationDate":"2017-06-14T17:58:29.504Z","title":"Assisted Feature Engineering and Feature Learning to Build Knowledge-Based Agents for Arcade Games","author_short":["Andelefski, B.","Schiffer, S."],"year":2017,"bibtype":"inproceedings","biburl":"https://kbsg.rwth-aachen.de/files/kbsgweb.bib","bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Bastian"],"propositions":[],"lastnames":["Andelefski"],"suffixes":[]},{"firstnames":["Stefan"],"propositions":[],"lastnames":["Schiffer"],"suffixes":[]}],"title":"Assisted Feature Engineering and Feature Learning to Build Knowledge-Based Agents for Arcade Games","year":"2017","pages":"228–238","editor":[{"firstnames":["Jaap"],"propositions":["van","den"],"lastnames":["Herik"],"suffixes":[]},{"firstnames":["Ana","Paula"],"propositions":[],"lastnames":["Rocha"],"suffixes":[]},{"firstnames":["Joaquim"],"propositions":[],"lastnames":["Filipe"],"suffixes":[]}],"booktitle":"ICAART 2017 - Proceedings of the 4th International Conference on Agents and Artificial Intelligence","volume":"2 - Artificial Intelligence","location":"Porto, Portugal","month":"February 6-8","publisher":"SciTePress","isbn":"978-989-758-220-2","abstract":"Human knowledge can greatly increase the performance of autonomous agents. Leveraging this knowledge is sometimes neither straightforward nor easy. In this paper, we present an approach for assisted feature engineering and feature learning to build knowledge-based agents for three arcade games within the Arcade Learning Environment. While existing approaches mostly use model-free approaches we aim at creating a descriptive set of features for world modelling and building agents. To this end, we provide (visual) assistance in identifying and modelling features from RAM, we allow for learning features based on labeled game data, and we allow for creating basic agents using the above features. In our evaluation, we compare different methods to learn features from the RAM. We then compare several agents using different sets of manual and learned features with one another and with the state-of-the-art.","bibtex":"@inproceedings{ Andelefski:Schiffer:ICAART2017:AssistedFeatureLearning,\n author = {Bastian Andelefski and Stefan Schiffer},\n title = {Assisted Feature Engineering and Feature Learning to Build Knowledge-Based Agents for Arcade Games},\n year = {2017},\n pages = {228--238},\n editor = {Jaap van den Herik and Ana Paula Rocha and Joaquim Filipe},\n booktitle = {ICAART 2017 - Proceedings of the 4th International Conference on Agents and Artificial Intelligence},\n volume = {2 - Artificial Intelligence},\n location = {Porto, Portugal},\n month = {February 6-8},\n publisher = {SciTePress},\n isbn = {978-989-758-220-2},\n abstract = {Human knowledge can greatly increase the performance\n of autonomous agents. Leveraging this knowledge is\n sometimes neither straightforward nor easy. In this\n paper, we present an approach for assisted feature\n engineering and feature learning to build\n knowledge-based agents for three arcade games within\n the Arcade Learning Environment. While existing\n approaches mostly use model-free approaches we aim\n at creating a descriptive set of features for world\n modelling and building agents. To this end, we\n provide (visual) assistance in identifying and\n modelling features from RAM, we allow for learning\n features based on labeled game data, and we allow\n for creating basic agents using the above\n features. In our evaluation, we compare different\n methods to learn features from the RAM. We then\n compare several agents using different sets of\n manual and learned features with one another and\n with the state-of-the-art.},\n}\n\n\n","author_short":["Andelefski, B.","Schiffer, S."],"editor_short":["van den Herik, J.","Rocha, A. P.","Filipe, J."],"key":"Andelefski:Schiffer:ICAART2017:AssistedFeatureLearning","id":"Andelefski:Schiffer:ICAART2017:AssistedFeatureLearning","bibbaseid":"andelefski-schiffer-assistedfeatureengineeringandfeaturelearningtobuildknowledgebasedagentsforarcadegames-2017","role":"author","urls":{},"metadata":{"authorlinks":{"schiffer, s":"https://kbsg.rwth-aachen.de/~schiffer/"}},"downloads":0},"search_terms":["assisted","feature","engineering","feature","learning","build","knowledge","based","agents","arcade","games","andelefski","schiffer"],"keywords":[],"authorIDs":["2szSuFfs9MafNRd67","4xuwkAN7nPCAMNzHf","54576ff92abc8e9f370003d0","58dfynaKtAENT9cH4","5de8996b2c5eb2df010000de","5de8cf119e80cdde01000138","5dea2854b5dcc6df01000199","5df7c58f92a8e4df01000119","5dfbf8574d4fb9de010001a3","5dffa5e6f16465df0100014f","5e159bbce8bfcff20100005e","5e183fe2bb35f5de01000077","5e1c5b5be556c6de0100016f","5e1ecc4f52ab76df01000171","5e20809180350bdf01000115","5e20a5b7bdda1fde01000179","5e29965efed3e7df0100004f","5e2ac0388319d5de01000156","5e2ee9e69033bede0100017f","5e3141635a3ceade0100014c","5e31563f8cf138de01000107","5e40172917f17dde0100029a","5e4262ff82b0c2de010000ff","5e44383be5a34dde0100015b","5e483ce700973ede01000025","5e4cca760cc7d3de01000086","5e50584cc708b1de0100012b","5e5670b4df3460df010000b4","5e5d3b7573eb2edf01000086","5e6a8f360e8744de01000102","6Jh3syL98hHpSWu6S","6KYm6sthg82iABu6p","6Q5igSDyKtjnw72as","DWc8C5RMKAEm6W5wx","FaTYSsvesqJLKtPef","Gcs9RCsvh8Hyn97ZN","PFJZ67Ahav7GY9N29","SqZqoeBZqnbSNdwxe","Th9jbA89wFSmvWBcc","a2Gv9Spw2CeaoyHEx","gKeviFc9xnzeQjzht","gMeNzvMudAEqjFiiX","nkgd6ekPnFftFxtN7","pAs8oLbgt7aaf9kAA","qkGZEsHcoggnbMpm4","spcssCfGoBC2pvSN5","vf22cpDWZrYX4x4w3","xezSR9ixeJATYXuiC","ywJqbJM8Yh4t8YkQ3"],"dataSources":["h8Pz4jiwoLbb3hmG2","dqRQPSg6Hy3ZXQg7z"]}