Timezone: »

 
Poster
What Can Linear Interpolation of Neural Network Loss Landscapes Tell Us?
Tiffany Vlaar · Jonathan Frankle

Wed Jul 20 03:30 PM -- 05:30 PM (PDT) @ Hall E #235
in Poster Session 2 »

Studying neural network loss landscapes provides insights into the nature of the underlying optimization problems. Unfortunately, loss landscapes are notoriously difficult to visualize in a human-comprehensible fashion. One common way to address this problem is to plot linear slices of the landscape, for example from the initial state of the network to the final state after optimization. On the basis of this analysis, prior work has drawn broader conclusions about the difficulty of the optimization problem. In this paper, we put inferences of this kind to the test, systematically evaluating how linear interpolation and final performance vary when altering the data, choice of initialization, and other optimizer and architecture design choices. Further, we use linear interpolation to study the role played by individual layers and substructures of the network. We find that certain layers are more sensitive to the choice of initialization, but that the shape of the linear path is not indicative of the changes in test accuracy of the model. Our results cast doubt on the broader intuition that the presence or absence of barriers when interpolating necessarily relates to the success of optimization.

Keywords: Deep Learning · Miscellaneous Aspects of Machine Learning
[ Slides]  [ Poster]  [ Paper PDF] 

Author Information

Tiffany Vlaar (University of Edinburgh)
Jonathan Frankle (MosaicML / Harvard)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors