Statistical Learning
A no-regret generalization of hierarchical softmax to extreme multi-label classification
Marek Wydmuch, Kalina Jasinska, Mikhail Kuznetsov, Róbert Busa-Fekete, Krzysztof Dembczynski
Learning problems of this scale are often referred to as extreme classification .
Learning to Optimize Tensor Programs
Tianqi Chen, Lianmin Zheng, Eddie Yan, Ziheng Jiang, Thierry Moreau, Luis Ceze, Carlos Guestrin, Arvind Krishnamurthy
Neural Information Processing Systems http://nips.cc/
Learning Confidence Sets using Support Vector Machines
Those points in the first two sets are classified to a single class as by traditional classifiers.
$\ell_1$-regression with Heavy-tailed Distributions
Linear regression used to be a mainstay of statistics, and remains one of our most important tools for data analysis [Hastie et al., 2009].
Watch Your Step: Learning Node Embeddings via Graph Attention
Sami Abu-El-Haija, Bryan Perozzi, Rami Al-Rfou, Alexander A. Alemi
Neural Information Processing Systems http://nips.cc/
Model-based targeted dimensionality reduction for neuronal population data
Neural Information Processing Systems http://nips.cc/
Stochastic Chebyshev Gradient Descent for Spectral Optimization
Insu Han, Haim Avron, Jinwoo Shin
Unfortunately, computing the gradient of a spectral function is generally of cubic complexity, as such gradient descent methods are rather expensive for optimizing objectives involving the spectral function.
How to tell when a clustering is (approximately) correct using convex relaxations
Neural Information Processing Systems http://nips.cc/
Benefits of over-parameterization with EM
Ji Xu, Daniel J. Hsu, Arian Maleki
Despite the fact that the weights fixed in Model 1, we now pretend that they are not fixed. This gives a second model, which we call Model 2 . Parameter estimation for Model 2 requires EM to estimate the mixing weights in addition to the component means.
Bayesian Distributed Stochastic Gradient Descent
Neural Information Processing Systems http://nips.cc/