%0 Conference Paper %1 z.gabor1998 %A Gábor, Z. %A Kalmár, Zs. %A Szepesvári, Cs. %B ICML %D 1998 %K MDPs constrained learning, reinforcement theory, %T Multi-criteria Reinforcement Learning %X We consider multi-criteria sequential decision making problems where the vector-valued evaluations are compared by a given, fixed total ordering. Conditions for the optimality of stationary policies and the Bellman optimality equation are given for a special, but important class of problems when the evaluation of policies can be computed for the criteria independently of each other. The analysis requires special care as the topology introduced by pointwise convergence and the order-topology introduced by the preference order are in general incompatible. Reinforcement learning algorithms are proposed and analyzed. Preliminary computer experiments confirm the validity of the derived algorithms. These type of multi-criteria problems are most useful when there are several optimal solutions to a problem and one wants to choose the one among these which is optimal according to another fixed criterion. Possible application in robotics and repeated games are outlined.

@inproceedings{z.gabor1998, abstract = {We consider multi-criteria sequential decision making problems where the vector-valued evaluations are compared by a given, fixed total ordering. Conditions for the optimality of stationary policies and the Bellman optimality equation are given for a special, but important class of problems when the evaluation of policies can be computed for the criteria independently of each other. The analysis requires special care as the topology introduced by pointwise convergence and the order-topology introduced by the preference order are in general incompatible. Reinforcement learning algorithms are proposed and analyzed. Preliminary computer experiments confirm the validity of the derived algorithms. These type of multi-criteria problems are most useful when there are several optimal solutions to a problem and one wants to choose the one among these which is optimal according to another fixed criterion. Possible application in robotics and repeated games are outlined.}, added-at = {2020-03-17T03:03:01.000+0100}, author = {G{\'a}bor, Z. and Kalm{\'a}r, {Zs}. and Szepesv{\'a}ri, {Cs}.}, biburl = {https://www.bibsonomy.org/bibtex/22c9419ffd8b86e995ee5b3184173997b/csaba}, booktitle = {ICML}, date-modified = {2010-09-02 13:09:16 -0600}, interhash = {00078c9690694eb9a56ca7866b5101c6}, intrahash = {2c9419ffd8b86e995ee5b3184173997b}, keywords = {MDPs constrained learning, reinforcement theory,}, note = {Revised on 05/04/2004}, owner = {Beata}, pdf = {papers/multi98.ps.pdf}, timestamp = {2020-03-17T03:03:01.000+0100}, title = {Multi-criteria Reinforcement Learning}, year = 1998 }