Web: http://arxiv.org/abs/2205.09615

Sept. 22, 2022, 1:12 a.m. | Ivan Karpukhin, Stanislav Dereka, Sergey Kolesnikov

cs.LG updates on arXiv.org arxiv.org

Classification tasks are usually evaluated in terms of accuracy. However,
accuracy is discontinuous and cannot be directly optimized using gradient
ascent. Popular methods minimize cross-entropy, hinge loss, or other surrogate
losses, which can lead to suboptimal results. In this paper, we propose a new
optimization framework by introducing stochasticity to a model's output and
optimizing expected accuracy, i.e. accuracy of the stochastic model. Extensive
experiments on linear models and deep image classification show that the
proposed optimization method is a …

accuracy arxiv

More from arxiv.org / cs.LG updates on arXiv.org

Postdoctoral Fellow: ML for autonomous materials discovery

@ Lawrence Berkeley National Lab | Berkeley, CA

Research Scientists

@ ODU Research Foundation | Norfolk, Virginia

Embedded Systems Engineer (Robotics)

@ Neo Cybernetica | Bedford, New Hampshire

2023 Luis J. Alvarez and Admiral Grace M. Hopper Postdoc Fellowship in Computing Sciences

@ Lawrence Berkeley National Lab | San Francisco, CA

Senior Manager Data Scientist

@ NAV | Remote, US

Senior AI Research Scientist

@ Earth Species Project | Remote anywhere

Research Fellow- Center for Security and Emerging Technology (Multiple Opportunities)

@ University of California Davis | Washington, DC

Staff Fellow - Data Scientist

@ U.S. FDA/Center for Devices and Radiological Health | Silver Spring, Maryland

Staff Fellow - Senior Data Engineer

@ U.S. FDA/Center for Devices and Radiological Health | Silver Spring, Maryland

Research Engineer - VFX, Neural Compositing

@ Flawless | Los Angeles, California, United States

[Job-TB] Senior Data Engineer

@ CI&T | Brazil

Data Analytics Engineer

@ The Fork | Paris, France