%0 Electronic Article %A Botvinick, Matthew and Barrett, David G. T. and Battaglia, Peter and de Freitas, Nando and Kumaran, Darshan and Leibo, Joel Z and Lillicrap, Timothy and Modayil, Joseph and Mohamed, Shakir and Rabinowitz, Neil C. and Rezende, Danilo J. and Santoro, Adam and Schaul, Tom and Summerfield, Christopher and Wayne, Greg and Weber, Theophane and Wierstra, Daan and Legg, Shane and Hassabis, Demis %I Cambridge University Press (CUP) %D 2017 %D 2017 %G English %@ 0140-525X %@ 1469-1825 %~ Hochschule Zittau / Görlitz, Hochschulbibliothek %T Building machines that learn and think for themselves %V 40 %J Behavioral and Brain Sciences %V 40 %U http://dx.doi.org/10.1017/s0140525x17000048 %X AbstractWe agree with Lake and colleagues on their list of “key ingredients” for building human-like intelligence, including the idea that model-based reasoning is essential. However, we favor an approach that centers on one additional ingredient: autonomy. In particular, we aim toward agents that can both build and exploit their own internal models, with minimal human hand engineering. We believe an approach centered on autonomous learning has the greatest chance of success as we scale toward real-world complexity, tackling domains for which ready-made formal models are not available. Here, we survey several important examples of the progress that has been made toward building autonomous agents with human-like abilities, and highlight some outstanding challenges. %Z https://katalog.hszg.de/Record/ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMTAxNy9zMDE0MDUyNXgxNzAwMDA0OA %U https://katalog.hszg.de/Record/ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMTAxNy9zMDE0MDUyNXgxNzAwMDA0OA