Timezone: »
Subsampling methods have been recently proposed to speed up least squares estimation in large scale settings. However, these algorithms are typically not robust to outliers or corruptions in the observed covariates. The concept of influence that was developed for regression diagnostics can be used to detect such corrupted observations as shown in this paper. This property of influence -- for which we also develop a randomized approximation -- motivates our proposed subsampling algorithm for large scale corrupted linear regression which limits the influence of data points since highly influential points contribute most to the residual error. Under a general model of corrupted observations, we show theoretically and empirically on a variety of simulated and real datasets that our algorithm improves over the current state-of-the-art approximation schemes for ordinary least squares.
Author Information
Brian McWilliams (DeepMind)
Gabriel Krummenacher (ETH Zurich)
Mario Lucic (Google Brain (Zurich))
Joachim M Buhmann (ETH Zurich)
Related Events (a corresponding poster, oral, or spotlight)
-
2014 Poster: Fast and Robust Least Squares Estimation in Corrupted Linear Models »
Thu. Dec 11th 12:00 -- 04:59 AM Room Level 2, room 210D
More from the Same Authors
-
2023 Poster: Explore In-Context Learning for 3D Point Cloud Understanding »
zhongbin fang · Xiangtai Li · Xia Li · Joachim M Buhmann · Chen Change Loy · Mengyuan Liu -
2023 Poster: Invariant Anomaly Detection under Distribution Shifts: A Causal Perspective »
João Carvalho · Mengtao Zhang · Robin Geyer · Carlos Cotrini · Joachim M Buhmann -
2022 Poster: Learning to Drop Out: An Adversarial Approach to Training Sequence VAEs »
Djordje Miladinovic · Kumar Shridhar · Kushal Jain · Max Paulus · Joachim M Buhmann · Carl Allen -
2022 Poster: Learning Long-Term Crop Management Strategies with CyclesGym »
Matteo Turchetta · Luca Corinzia · Scott Sussex · Amanda Burton · Juan Herrera · Ioannis Athanasiadis · Joachim M Buhmann · Andreas Krause -
2017 Poster: Efficient and Flexible Inference for Stochastic Systems »
Stefan Bauer · Nico S Gorbach · Djordje Miladinovic · Joachim M Buhmann -
2017 Poster: Non-monotone Continuous DR-submodular Maximization: Structure and Algorithms »
Yatao Bian · Kfir Levy · Andreas Krause · Joachim M Buhmann -
2017 Poster: Scalable Variational Inference for Dynamical Systems »
Nico S Gorbach · Stefan Bauer · Joachim M Buhmann -
2017 Poster: Stochastic Submodular Maximization: The Case of Coverage Functions »
Mohammad Karimi · Mario Lucic · Hamed Hassani · Andreas Krause -
2016 Poster: Scalable Adaptive Stochastic Optimization Using Random Projections »
Gabriel Krummenacher · Brian McWilliams · Yannic Kilcher · Joachim M Buhmann · Nicolai Meinshausen -
2016 Poster: Fast and Provably Good Seedings for k-Means »
Olivier Bachem · Mario Lucic · Hamed Hassani · Andreas Krause -
2016 Oral: Fast and Provably Good Seedings for k-Means »
Olivier Bachem · Mario Lucic · Hamed Hassani · Andreas Krause -
2015 Poster: Variance Reduced Stochastic Gradient Descent with Neighbors »
Thomas Hofmann · Aurelien Lucchi · Simon Lacoste-Julien · Brian McWilliams -
2013 Poster: Correlated random features for fast semi-supervised learning »
Brian McWilliams · David Balduzzi · Joachim M Buhmann -
2011 Workshop: Philosophy and Machine Learning »
Marcello Pelillo · Joachim M Buhmann · Tiberio Caetano · Bernhard Schölkopf · Larry Wasserman -
2006 Poster: Denoising and Dimension Reduction in Feature Space »
Mikio L Braun · Joachim M Buhmann · Klaus-Robert Müller