Online metric algorithms with untrusted predictions
Jul 12, 2020
Speakers
Antonios Antoniadis
Sprecher:in · 0 Follower:innen
Christian Coester
Sprecher:in · 0 Follower:innen
Marek Eliáš
Sprecher:in · 0 Follower:innen
Adam Polak
Sprecher:in · 0 Follower:innen
Bertrand Simon
Sprecher:in · 0 Follower:innen
About
Machine-learned predictors, although achieving very good results for inputs resembling training data, cannot possibly provide perfect predictions in all situations. Still, decision-making systems that are based on such predictors need not only to benefit from good predictions but also to achieve a decent performance when the predictions are inadequate. In this paper, we propose a prediction setup for Metrical Task Systems (MTS), a broad class of online decision-making problems including, e.g., caching, k-server and convex body chasing. We utilize results from the theory of online algorithms to show how to make the setup robust. We extend our setup in two ways, (1) adapting it beyond MTS to the online matching on the line problem, and (2) specifically for caching, slightly enriching the predictor’s output to achieve an improved dependence on the prediction error. Finally, we present an empirical evaluation of our methods on real world datasets, which suggests practicality.Machine-learned predictors, although achieving very good results for inputs resembling training data, cannot possibly provide perfect predictions in all situations. Still, decision-making systems that are based on such predictors need not only to benefit from good predictions but also to achieve a decent performance when the predictions are inadequate. In this paper, we propose a prediction setup for Metrical Task Systems (MTS), a broad class of online decision-making problems including, e.g., c…
Categories
Informatik und IT
Kategorie · 14,8k Präsentationen
About ICML 2020
The International Conference on Machine Learning (ICML) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence known as machine learning. ICML is globally renowned for presenting and publishing cutting-edge research on all aspects of machine learning used in closely related areas like artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, and robotics. ICML is one of the fastest growing artificial intelligence conferences in the world. Participants at ICML span a wide range of backgrounds, from academic and industrial researchers, to entrepreneurs and engineers, to graduate students and postdocs.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Invited Talk 3
Später ansehen
Bayesian Optimization for Iterative Learning
Später ansehen
Vu Nguyen, …
Baidu PaddlePaddle – An industry-grade end-to-end deep learning platform
Später ansehen
Empirical Study of the Benefits of Overparameterization in Learning Latent Variable Models
Später ansehen
