Statistical Learning
The Cells Out of Sample (COOS) dataset and benchmarks for measuring out-of-sample generalization of image classifiers
Alex Lu, Amy Lu, Wiebke Schormann, Marzyeh Ghassemi, David Andrews, Alan Moses
Neural Information Processing Systems http://nips.cc/
Fair Algorithms for Clustering
Suman Bera, Deeparnab Chakrabarty, Nicolas Flores, Maryam Negahbani
We allow the user to specify the parameters that define fair representation.
Online Convex Matrix Factorization with Representative Regions
Jianhao Peng, Olgica Milenkovic, Abhishek Agarwal
Matrix factorization (MF) is a versatile learning method that has found wide applications in various data-driven disciplines.
while A.3 is a common assumption in linear regression analysis and it relates to the LARS problem in our online
We thank all the reviewers for carefully reading of the manuscript and constructive comments. Reviewer #1: Assumptions A.2 and A.3 used in Algorithm 2. Reviewer #3: There seem to be several misunderstandings regarding the steps of our algorithm and its analysis. To maintain the list constant, for every added point another point is removed.
Who is Afraid of Big Bad Minima? Analysis of gradient-flow in spiked matrix-tensor models
Stefano Sarao Mannelli, Giulio Biroli, Chiara Cammarota, Florent Krzakala, Lenka Zdeborová
When the signal-to-noise ratio is large enough, the success of gradient descent can thus be understood by a trivialisation transition in the loss landscape: either there is only a single minima, or all minima become "good", and no
Primal-Dual Block Generalized Frank-Wolfe Qi Lei
We propose a Frank-Wolfe (FW) type method that attains these three goals.
On the Downstream Performance of Compressed Word Embeddings
Avner May, Jian Zhang, Tri Dao, Christopher Ré
Neural Information Processing Systems http://nips.cc/
KNG: The K-Norm Gradient Mechanism
Neural Information Processing Systems http://nips.cc/
Efficient online learning with Kernels for adversarial large scale problems
We are interested in a framework of online learning with kernels for low-dimensional, but large-scale and potentially adversarial datasets.
Neural Diffusion Distance for Image Segmentation
The network is a differentiable deep architecture consisting of feature extraction and diffusion distance modules for computing diffusion distance on image by end-to-end training. We design low resolution kernel matching loss and high resolution segment matching loss to enforce the network's output to be consistent with human-labeled image segments.