Timezone: »
In most machine learning training paradigms a fixed, often handcrafted, loss function is assumed to be a good proxy for an underlying evaluation metric. In this work we assess this assumption by meta-learning an adaptive loss function to directly optimize the evaluation metric. We propose a sample efficient reinforcement learning approach for adapting the loss dynamically during training. We empirically show how this formulation improves performance by simultaneously optimizing the evaluation metric and smoothing the loss landscape. We verify our method in metric learning and classification scenarios, showing considerable improvements over the state-of-the-art on a diverse set of tasks. Importantly, our method is applicable to a wide range of loss functions and evaluation metrics. Furthermore, the learned policies are transferable across tasks and data, demonstrating the versatility of the method.
Author Information
Chen Huang (Apple Inc.)
Shuangfei Zhai (Apple)
Walter Talbott (Apple)
Miguel Angel Bautista Martin (Apple Inc.)
Shih-Yu Sun (Apple)
Carlos Guestrin (Apple & Univesity of Washington)
Josh Susskind (Apple, Inc.)
Related Events (a corresponding poster, oral, or spotlight)
-
2019 Oral: Addressing the Loss-Metric Mismatch with Adaptive Loss Alignment »
Wed Jun 12th 11:25 -- 11:30 PM Room Hall A
More from the Same Authors
-
2020 Poster: Equivariant Neural Rendering »
Emilien Dupont · Miguel Angel Bautista Martin · Alex Colburn · Aditya Sankar · Joshua Susskind · Qi Shan -
2020 Poster: AdaScale SGD: A User-Friendly Algorithm for Distributed Training »
Tyler Johnson · Pulkit Agrawal · Haijie Gu · Carlos Guestrin