Neural Information Processing Systems http://nips.cc/

In machine learning, many fundamental quantities we care to optimize such as entropy, graph cuts, diversity, coverage, diffusion, and clustering are submodular functions.

Neural network training relies on our ability to find "good" minimizers of highly

Arsenal's 10-game winning run was ended by Sunderland before the international break, but will the Premier League leaders bounce back against north London rivals Tottenham on Sunday? I feel like Arsenal are miles ahead of Tottenham at the moment but Spurs boss Thomas Frank will have a plan for this game, said BBC Sport football expert Chris Sutton. Frank will try to nullify Arsenal, who have not always been at their scintillating best this season - but Mikel Arteta's side usually still find a way to win. Sutton is making predictions for all 380 Premier League games this season, against AI, BBC Sport readers and a variety of guests. For week 12, he takes on singer Olly Murs, who is a Manchester United fan . Murs' new album, Knees Up, is out on Friday.

In current BCIs, the stimulus presentation schedule is typically generated in a pseudo-random fashion. Given the non-stationarity of brain electrical signals, a better strategy could be to adapt the stimulus presentation schedule in real-time by selecting the optimal stimuli that will maximize the signal-to-noise ratios of the elicited neural responses and provide the most information about the user's intent based on the uncertainties of the data being measured. However, the high-dimensional stimulus space limits the development of algorithms with tractable solutions for optimized stimulus selection to allow for real-time decision-making within the stringent time requirements of BCI processing.

Neural Information Processing Systems http://nips.cc/

One approach might be to capture the important structure of the current environment in a maximally compact way (to preserve capacity for future learning). Such learning is likely to result in positive transfer if future training domains share some structural similarity with the old ones.

These deployments have revealed gaps between existing theory and the needs of practitioners.

We present an end-to-end framework for solving the V ehicle Routing Problem (VRP) using reinforcement learning.

We propose a parsimonious quantile regression framework to learn the dynamic tail behaviors of financial asset returns.