Our natural gradient approach enables application of parallel filtering andsmoothing, further reducing thetemporal spancomplexitytobelogarithmic inthe number oftime steps.

Visual Reinforcement Learning (RL) agents trained on limited views face significant challenges in generalizing their learned abilities to unseen views.

Wefocusonsixmethods:(i)discriminative K-means (DisKmeans) in Ye et al. (2008); (ii) a discriminative clustering formulation described inBach andHarchaoui (2008); Flammarion etal.(2017); We compare two classesF of feature mappings: linear functions and fully-connected neural networks with one hidden layer that has 100 nodes. An epoch refers ton/B = 12 consecutive iterations. The learning curves in Figure 1 shows the advantage of neural network and demonstrates the flexibility of CURE with nonlinear function classes. One of the main obstacles is the complicated piecewise definition off, which prevent us from obtaining closed form formulae.