4 papers
Bootstrapped Representation Learning for Skeleton-Based Action Recognition
Olivier Moliner, Sangxia Huang, Kalle Åström
In this work, we study self-supervised representation learning for 3D skeleton-based action recognition. We extend Bootstrap Your Own Latent (BYOL) for representation learning on s…
Adversarial Attacks and Defences Competition
Alexey Kurakin, Ian Goodfellow, Samy Bengio +20
To accelerate research on adversarial examples and robustness of machine learning classifiers, Google Brain organized a NIPS 2017 competition that encouraged researchers to develop…
Semi-Supervised Algorithms for Approximately Optimal and Accurate Clustering
Buddhima Gamlath, Sangxia Huang, Ola Svensson
We study -means clustering in a semi-supervised setting. Given an oracle that returns whether two given points belong to the same cluster in a fixed optimal clustering, we inves…
Improved Hardness of Approximating Chromatic Number
Sangxia Huang
We prove that for sufficiently large K, it is NP-hard to color K-colorable graphs with less than 2^{K^{1/3}} colors. This improves the previous result of K versus K^{O(log K)} in K…