ys when planning meant searching for a sequence of abstract actions that satisfied some symbolic predicate. Robots can now learn their own representations through statistical infe...
ctural abstractions that were introduced in [17]. Compared with other symmetry-breaking techniques, the big advantage of dynamic symmetry breaking is that it can accommodate dynami...
Daniel S. Heller, Aurojit Panda, Meinolf Sellmann,...
In many recent object recognition systems, feature extraction
stages are generally composed of a filter bank, a
non-linear transformation, and some sort of feature pooling
layer...
Kevin Jarrett, Koray Kavukcuoglu, Marc’Aurelio R...
The new design for the introductory programming course at ETH relies on object technology, Eiffel, extensive reuse, a graphics-rich library (TRAFFIC) built specifically for the c...
We study the problemof statisticallycorrect inference in networks whose basic representations are population codes. Population codes are ubiquitous in the brain, and involve the s...