Robustness of edited neural networks

by   Davis Brown, et al.

Successful deployment in uncertain, real-world environments requires that deep learning models can be efficiently and reliably modified in order to adapt to unexpected issues. However, the current trend toward ever-larger models makes standard retraining procedures an ever-more expensive burden. For this reason, there is growing interest in model editing, which enables computationally inexpensive, interpretable, post-hoc model modifications. While many model editing techniques are promising, research on the properties of edited models is largely limited to evaluation of validation accuracy. The robustness of edited models is an important and yet mostly unexplored topic. In this paper, we employ recently developed techniques from the field of deep learning robustness to investigate both how model editing affects the general robustness of a model, as well as the robustness of the specific behavior targeted by the edit. We find that edits tend to reduce general robustness, but that the degree of degradation depends on the editing algorithm chosen. In particular, robustness is best preserved by more constrained techniques that modify less of the model. Motivated by these observations, we introduce two new model editing algorithms, direct low-rank model editing and 1-layer interpolation (1-LI), which each exhibit strong generalization performance.


Explicit Image Caption Editing

Given an image and a reference caption, the image caption editing task a...

A Revisit of Shape Editing Techniques: from the Geometric to the Neural Viewpoint

3D shape editing is widely used in a range of applications such as movie...

Editing Large Language Models: Problems, Methods, and Opportunities

Recent advancements in deep learning have precipitated the emergence of ...

Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptors

Large pre-trained models decay over long-term deployment as input distri...

Fast Model Editing at Scale

While large pre-trained models have enabled impressive results on a vari...

LIG-CRIStAL System for the WMT17 Automatic Post-Editing Task

This paper presents the LIG-CRIStAL submission to the shared Automatic P...

Latexify Math: Mathematical Formula Markup Revision to Assist Collaborative Editing in Math Q A Sites

Collaborative editing questions and answers plays an important role in q...

Please sign up or login with your details

Forgot password? Click here to reset