[{"_id":"19523","language":[{"iso":"eng"}],"year":"2019","author":[{"last_name":"Pfannschmidt","full_name":"Pfannschmidt, Karlson","first_name":"Karlson"},{"first_name":"Pritha","last_name":"Gupta","full_name":"Gupta, Pritha"},{"full_name":"Hüllermeier, Eyke","last_name":"Hüllermeier","first_name":"Eyke"}],"citation":{"short":"K. Pfannschmidt, P. Gupta, E. Hüllermeier, ArXiv:1901.10860 (2019).","chicago":"Pfannschmidt, Karlson, Pritha Gupta, and Eyke Hüllermeier. “Learning Choice Functions: Concepts and Architectures.” *ArXiv:1901.10860*, 2019.","ama":"Pfannschmidt K, Gupta P, Hüllermeier E. Learning Choice Functions: Concepts and Architectures. *arXiv:190110860*. 2019.","mla":"Pfannschmidt, Karlson, et al. “Learning Choice Functions: Concepts and Architectures.” *ArXiv:1901.10860*, 2019.","apa":"Pfannschmidt, K., Gupta, P., & Hüllermeier, E. (2019). Learning Choice Functions: Concepts and Architectures. *ArXiv:1901.10860*.","bibtex":"@article{Pfannschmidt_Gupta_Hüllermeier_2019, title={Learning Choice Functions: Concepts and Architectures}, journal={arXiv:1901.10860}, author={Pfannschmidt, Karlson and Gupta, Pritha and Hüllermeier, Eyke}, year={2019} }","ieee":"K. Pfannschmidt, P. Gupta, and E. Hüllermeier, “Learning Choice Functions: Concepts and Architectures,” *arXiv:1901.10860*. 2019."},"user_id":"13472","status":"public","date_created":"2020-09-17T10:53:38Z","abstract":[{"lang":"eng","text":"We study the problem of learning choice functions, which play an important\r\nrole in various domains of application, most notably in the field of economics.\r\nFormally, a choice function is a mapping from sets to sets: Given a set of\r\nchoice alternatives as input, a choice function identifies a subset of most\r\npreferred elements. Learning choice functions from suitable training data comes\r\nwith a number of challenges. For example, the sets provided as input and the\r\nsubsets produced as output can be of any size. Moreover, since the order in\r\nwhich alternatives are presented is irrelevant, a choice function should be\r\nsymmetric. Perhaps most importantly, choice functions are naturally\r\ncontext-dependent, in the sense that the preference in favor of an alternative\r\nmay depend on what other options are available. We formalize the problem of\r\nlearning choice functions and present two general approaches based on two\r\nrepresentations of context-dependent utility functions. Both approaches are\r\ninstantiated by means of appropriate neural network architectures, and their\r\nperformance is demonstrated on suitable benchmark tasks."}],"publication":"arXiv:1901.10860","date_updated":"2022-01-06T06:54:06Z","project":[{"_id":"52","name":"Computing Resources Provided by the Paderborn Center for Parallel Computing"}],"title":"Learning Choice Functions: Concepts and Architectures","department":[{"_id":"7"},{"_id":"355"}],"type":"preprint"}]