Improving Simple Models with Confidence Profiles

By Amit Dhurandhar, Karthikeyan Shanmugam, Ronny Luss, Peder A. Olsen. 2018

49 citations recorded by the source corpus.

In this paper, we propose a new method called ProfWeight for transferring information from a pre-trained deep neural network that has a high test accuracy to a simpler interpretable model or a very shallow network of low complexity and a priori low test accuracy. We are motivated by applications in interpretability and model deployment in severely memory constrained environments (like sensors). Our method uses linear probes to generate confidence scores through flattened intermediate representations. Our transfer method involves a theoretically justified weighting of samples during the training of the simple model using confidence scores of these intermediate layers. The value of our method is first demonstrated on CIFAR-10, where our weighting method significantly improves (3-4%) networks with only a fraction of the number of Resnet blocks of a complex Resnet model. We further demonstrate operationally significant results on a real manufacturing problem, where we dramatically increase the test accuracy of a CART model (the domain standard) by roughly 13%.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Explainable machine learning in deployment (2019, background): “These techniques attempt to characterize the concepts learned by the model [31], simpler models learned from the representation of complex models [17], prototypical samples from a particular model output [10], or the topology of the data itself [20].”
One Explanation Does Not Fit All: A Toolkit and Taxonomy of AI Explainability Techniques (2019, background): “• ProfWeight [20]: Learns a reweighting of the training set based on a given interpretable model and a high-performing complex neural network.”
Human-Centered Explainable AI (XAI): From Algorithms to User Experiences (2021, background): “How (global model-wide) •Describe the general model logic as feature impact∗, rules† or decisiontrees‡ • If user is only interested in a high-level view, describe what are the top features or rules considered ProfWeight∗†‡ [28], Global feature importance∗ [69, 102], Global feature inspection plots∗ ”
Explainable AI: A Review of Machine Learning Interpretability Methods (2020, background): “[63] proposed transferring knowledge from highperforming pre-trained deep neural networks to a low performing, but interpretable non-complex model to improve its performance.”