TY - GEN
T1 - Consistency of feature Markov processes
AU - Sunehag, Peter
AU - Hutter, Marcus
PY - 2010
Y1 - 2010
N2 - We are studying long term sequence prediction (forecasting). We approach this by investigating criteria for choosing a compact useful state representation. The state is supposed to summarize useful information from the history. We want a method that is asymptotically consistent in the sense it will provably eventually only choose between alternatives that satisfy an optimality property related to the used criterion. We extend our work to the case where there is side information that one can take advantage of and, furthermore, we briefly discuss the active setting where an agent takes actions to achieve desirable outcomes.
AB - We are studying long term sequence prediction (forecasting). We approach this by investigating criteria for choosing a compact useful state representation. The state is supposed to summarize useful information from the history. We want a method that is asymptotically consistent in the sense it will provably eventually only choose between alternatives that satisfy an optimality property related to the used criterion. We extend our work to the case where there is side information that one can take advantage of and, furthermore, we briefly discuss the active setting where an agent takes actions to achieve desirable outcomes.
UR - http://www.scopus.com/inward/record.url?scp=78249250264&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-16108-7_29
DO - 10.1007/978-3-642-16108-7_29
M3 - Conference contribution
SN - 3642161073
SN - 9783642161070
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 360
EP - 374
BT - Algorithmic Learning Theory - 21st International Conference, ALT 2010, Proceedings
T2 - 21st International Conference on Algorithmic Learning Theory, ALT 2010
Y2 - 6 October 2010 through 8 October 2010
ER -