TY - GEN
T1 - Testing and learning on distributions with symmetric noise invariance
AU - Law, Ho Chung Leon
AU - Yau, Christopher
AU - Sejdinovic, Dino
PY - 2017/1/1
Y1 - 2017/1/1
N2 - Kernel embeddings of distributions and the Maximum Mean Discrepancy (MMD), the resulting distance between distributions, are useful tools for fully nonparametric two-sample testing and learning on distributions. However, it is rare that all possible differences between samples are of interest - discovered differences can be due to different types of measurement noise, data collection artefacts or other irrelevant sources of variability. We propose distances between distributions which encode invariance to additive symmetric noise, aimed at testing whether the assumed true underlying processes differ. Moreover, we construct invariant features of distributions, leading to learning algorithms robust to the impairment of the input distributions with symmetric additive noise.
AB - Kernel embeddings of distributions and the Maximum Mean Discrepancy (MMD), the resulting distance between distributions, are useful tools for fully nonparametric two-sample testing and learning on distributions. However, it is rare that all possible differences between samples are of interest - discovered differences can be due to different types of measurement noise, data collection artefacts or other irrelevant sources of variability. We propose distances between distributions which encode invariance to additive symmetric noise, aimed at testing whether the assumed true underlying processes differ. Moreover, we construct invariant features of distributions, leading to learning algorithms robust to the impairment of the input distributions with symmetric additive noise.
UR - https://www.scopus.com/pages/publications/85046992774
M3 - Conference contribution
AN - SCOPUS:85046992774
T3 - Advances in Neural Information Processing Systems
SP - 1344
EP - 1354
BT - Advances in Neural Information Processing Systems 30 (NIPS 2017)
A2 - Guyon, I.
A2 - Von Luxburg, U.
A2 - Bengio, S.
A2 - Wallach, H.
A2 - Fergus, R.
A2 - Vishwanathan, S.
A2 - Garnett, R.
PB - NeurIPS
T2 - 31st Annual Conference on Neural Information Processing Systems, NIPS 2017
Y2 - 4 December 2017 through 9 December 2017
ER -