Filters








11 Hits in 7.9 sec

Local Perturb-and-MAP for Structured Prediction [article]

Gedas Bertasius, Qiang Liu, Lorenzo Torresani, Jianbo Shi
2016 arXiv   pre-print
of the original CRF model.  ...  Conditional random fields (CRFs) provide a powerful tool for structured prediction, but cast significant challenges in both the learning and inference steps.  ...  the 29th International Conference on Machine Learning, ICML 2012, Edinburgh, Scotland, UK, June 26 - July 1, 2012, 2012.  ... 
arXiv:1605.07686v2 fatcat:vgezpek7n5eerid26lkb7hx56u

Worst Cases Policy Gradients [article]

Yichuan Charlie Tang, Jian Zhang, Ruslan Salakhutdinov
2019 arXiv   pre-print
The learned policy can map the same state to different actions depending on the propensity for risk.  ...  Towards this goal, we propose an actor-critic framework that models the uncertainty of the future and simultaneously learns a policy based on that uncertainty model.  ...  In Proceedings of the 29th International Conference on Machine Learning, ICML 2012, Edinburgh, Scotland, UK, June 26 - July 1, 2012, 2012. URL http://icml.cc/2012/papers/489.pdf. [42] M. L.  ... 
arXiv:1911.03618v1 fatcat:allc4pk3qzburak6jg6rcmnqdy

Deep Spoken Keyword Spotting: An Overview

Ivan Lopez-Espejo, Zheng-Hua Tan, John Hansen, Jesper Jensen
2021 IEEE Access  
Rigoll, “Keyword spotting exploiting Learning, June 26-July 1, Edinburgh, Scotland, 2012.  ...  Proceedings of ICML 201229th International Conference on Machine [84] M. Wöllmer, B. Schuller, and G.  ... 
doi:10.1109/access.2021.3139508 fatcat:i4pfpfxcpretlkbefp7owtxcti

Algorithmic Fairness Datasets: the Story so Far [article]

Alessandro Fabris, Stefano Messina, Gianmaria Silvello, Gian Antonio Susto
2022 arXiv   pre-print
Progress in fair Machine Learning hinges on data, which can be appropriately used only if adequately documented.  ...  As a result, a growing community of researchers has been investigating the equity of existing algorithms and proposing novel ones, advancing the understanding of risks and opportunities of automated decision-making  ...  Acknowledgements The authors would like to thank the following researchers and dataset creators for the useful feedback on the data briefs: Alain Barrat, Luc Behaghel, Asia Biega, Marko Bohanec, Chris  ... 
arXiv:2202.01711v2 fatcat:5hf4a42pubc5vnt7tw3al4m5bq

An Equivalence Between Data Poisoning and Byzantine Gradient Attacks [article]

Sadegh Farhadkhani, Rachid Guerraoui, Lê-Nguyên Hoang, Oscar Villemaud
2022 arXiv   pre-print
This equivalence makes it possible to obtain new impossibility results on the resilience to data poisoning as corollaries of existing impossibility theorems on Byzantine machine learning.  ...  To study the resilience of distributed learning, the "Byzantine" literature considers a strong threat model where workers can report arbitrary gradients to the parameter server.  ...  In Proceedings of the 29th International Conference on Machine Learning, ICML 2012, Edinburgh, Scotland, UK, June 26 -July 1, 2012. icml.cc / Omnipress, 2012. and optimal algorithms for personalized federated  ... 
arXiv:2202.08578v1 fatcat:arnkw6u5f5ecdblvxetfqwsccm

Convergence rate of stochastic k-means [article]

Cheng Tang, Claire Monteleoni
2016 arXiv   pre-print
The k-means objective is non-convex and non-differentiable: we exploit ideas from non-convex gradient-based optimization by providing a novel characterization of the trajectory of k-means algorithm on  ...  Both scale up the widely used Lloyd 's algorithm via stochastic approximation, and have become popular for large-scale clustering and unsupervised feature learning.  ...  In Proceedings of the 29th International Conference on Machine Learning, ICML 2012, Edinburgh, Scotland, UK, June 26 -July 1, 2012, 2012. 7 Appendix A: supplementary materials to Sec- tion 3.1  ... 
arXiv:1610.04900v2 fatcat:5zpfr3qxbjg4hebnbmgzud3y4e

A universal probabilistic spike count model reveals ongoing modulation of neural variability [article]

David Liu, Mate Lengyel
2021 bioRxiv   pre-print
In addition, they make strong assumptions about the parametric form of variability, rely on assumption-free but data-inefficient histogram-based approaches, or are altogether ill-suited for capturing variability  ...  Our approach paves the way to understanding the mechanisms and computations underlying neural variability under naturalistic conditions, beyond the realm of sensory coding with repeatable stimuli.  ...  Melkonyan for helpful feedback on the manuscript. The number of inducing points has been shown to scale favourably as O((log T ) D ) for standard Gaussian process regression models [40] .  ... 
doi:10.1101/2021.06.27.450063 fatcat:t3tohitynfamrnqrhbryj36owq

An Empirical Process Approach to the Union Bound: Practical Algorithms for Combinatorial and Linear Bandits [article]

Julian Katz-Samuels, Lalit Jain, Zohar Karnin, Kevin Jamieson
2020 arXiv   pre-print
G-optimal design), we define an experimental design objective based on the Gaussian-width of the underlying arm set.  ...  Leveraging ideas from the theory of suprema of empirical processes, we provide an algorithm whose sample complexity scales with the geometry of the instance and avoids an explicit union bound over the  ...  In International Conference on Artificial Intelligence and Statistics, Machine Learning, ICML 2012, Edinburgh, Scotland, UK, June 26 -July 1, 2012, 2012.  ... 
arXiv:2006.11685v1 fatcat:6q7vjyyncbgqdnjcenfincitry

Algorithmic Fairness Datasets: the Story so Far [article]

Alessandro Fabris, Stefano Messina, Gianmaria Silvello, Gian Antonio Susto
2022
As a result, a growing community of algorithmic fairness researchers has been investigating the equity of existing algorithms and proposing novel ones, advancing the understanding of the risks and opportunities  ...  Unfortunately, the algorithmic fairness community, as a whole, suffers from a collective data documentation debt caused by a lack of information on specific resources (opacity) and scatteredness of available  ...  Acknowledgements The authors would like to thank the following researchers and dataset creators for the useful feedback on the data briefs: Alain Barrat, Luc  ... 
doi:10.48550/arxiv.2202.01711 fatcat:mav36x3w5namjhurzpevtsmsju

Towards Optimal Algorithms for Prediction with Expert Advice [article]

Nick Gravin, Yuval Peres, Balasubramanian Sivan
2016 arXiv   pre-print
Our analysis of the optimal adversary goes through delicate asymptotics of the random walk of a particle between multiple walls.  ...  We study the classical problem of prediction with expert advice in the adversarial setting with a geometric stopping time. In 1965, Cover gave the optimal algorithm for the case of 2 experts.  ...  In Proceedings of the 29th International Conference on Machine Learning, ICML 2012, Edinburgh, Scotland, UK, June 26 -July 1, 2012, 2012. [9] Yoav Freund and Robert E. Schapire.  ... 
arXiv:1409.3040v5 fatcat:acpxysce55egxhaw32ihoi7jsy

Comparing graphs

Nils Morten Kriege, Technische Universität Dortmund, Technische Universität Dortmund
2015
We present algorithms with improved bounds on running time for the subclass of outerplanar graphs.  ...  These are a prerequisite for the application of a variety of data mining algorithms to the domain of graphs. Hence, various approaches to graph comparison evolved and are wide-spread in practice.  ...  In Proceedings of the 29th International Conference on Machine Learning, ICML 2012, Edinburgh, Scotland, UK, June 26 -July 1, 2012. icml.cc / Omnipress, 2012 • Nils Kriege, Marion Neumann, Kristian  ... 
doi:10.17877/de290r-16358 fatcat:btrj2zdwlngxncyr7gpywat4be