`

Timezone: »

 
Poster
Minimax Lower Bounds for Transfer Learning with Linear and One-hidden Layer Neural Networks
Mohammadreza Mousavi Kalan · Zalan Fabian · Salman Avestimehr · Mahdi Soltanolkotabi

Tue Dec 08 09:00 PM -- 11:00 PM (PST) @ Poster Session 2 #730
in Poster Session 3 »

Transfer learning has emerged as a powerful technique for improving the performance of machine learning models on new domains where labeled training data may be scarce. In this approach a model trained for a source task, where plenty of labeled training data is available, is used as a starting point for training a model on a related target task with only few labeled training data. Despite recent empirical success of transfer learning approaches, the benefits and fundamental limits of transfer learning are poorly understood. In this paper we develop a statistical minimax framework to characterize the fundamental limits of transfer learning in the context of regression with linear and one-hidden layer neural network models. Specifically, we derive a lower-bound for the target generalization error achievable by any algorithm as a function of the number of labeled source and target data as well as appropriate notions of similarity between the source and target tasks. Our lower bound provides new insights into the benefits and limitations of transfer learning. We further corroborate our theoretical finding with various experiments.

[ Paper ]

Author Information

Mohammadreza Mousavi Kalan (University of Southern California)
Zalan Fabian (University of Southern California)
Salman Avestimehr (University of Southern California)
Mahdi Soltanolkotabi (University of Southern california)

More from the Same Authors