184
click to vote
AGI
2015
9 years 12 months ago
2015
Abstract. Some currently popular and successful deep learning architectures display certain pathological behaviors (e.g. confidently classifying random data as belonging to a fami...
AGI
2015
9 years 12 months ago
2015
Abstract. Artificial emotions of different varieties have been used for controlling behavior, e.g. in cognitive architectures and reinforcement learning models. We propose to use ...
AGI
2015
9 years 12 months ago
2015
This paper motivates the study of counterpossibles (logically impossible counterfactuals) as necessary for developing a decision theory suitable for generally intelligent agents e...
106
click to vote
AGI
2015
9 years 12 months ago
2015
This paper argues the possibility of designing AI that can learn logics from data. We provide an abstract framework for learning logics. In this framework, an agent A provides trai...
AGI
2015
9 years 12 months ago
2015
We propose that Solomonoff induction is complete in the physical sense via several strong physical arguments. We also argue that Solomonoff induction is fully applicable to quant...