SPS Education Short Course 

Title:  Visual Explainability in Machine Learning

Presented by: Ghassan AlRegib, and Mohit Prabhushankar 

Omni Lab for Intelligent Visual Engineering and Science (OLIVES) 

School of Electrical and Computer Engineering  

Georgia Institute of Technology, Atlanta, USA 

https://alregib.ece.gatech.edu/

Watch the tutorial recording: HERE

Course Overview 

Visual explanations have traditionally acted as rationales used to justify the decisions made by machine learning systems. With the advent of large-scale neural networks, the role of visual explanations has been to shed interpretability on black-box models. We view this role as the process for the network to answer the question `Why P?’, where P is a trained network’s prediction. Recently however, with increasingly capable models, the role of explainability has expanded. Neural networks are asked to justify `What if?’ counterfactual and `Why P, rather than Q?’ contrastive question modalities that the network did not explicitly train to answer. This allows explanations to act as reasons to make further prediction. The short course provides a principled and rational introduction into Explainability within machine learning and justifies them as reasons to make decisions. Such a reasoning framework allows for robust machine learning as well as trustworthy AI to be accepted in everyday lives. Applications like robust recognition, image quality assessment, visual saliency, anomaly detection, out-of-distribution detection, adversarial image detection, seismic interpretation, semantic segmentation, and machine teaching among others will be discussed. 

Learning Outcome

• Impress on the importance of Explainability in AI systems as a function of humans (users, engineers, researchers, and policymakers) requiring it
• Define Explainability and characterize it based on its required properties, methodologies and the intended audience it caters to
• Detail popular visual explanatory techniques across multiple data modalities including natural images, biomedical and seismic images, and videos
• Expand on subjective and objective techniques to evaluate explanations
• Discuss accepted proxies for Explainability – robustness and uncertainty
• Contrast against data-specific instantiations of Explainability
• Consider alternative data and explanation-centric training regimen
• Debate on the role of Visual Explainability through the lens of causality and Generative AI

Schedule

Day 1 (Tuesday, 12/05)

Day 2 (Wednesday, 12/06)

Day 3 (Thursday, 12/07)

Target audience

The targeted audiences are senior-year undergraduate, postgraduate, engineers and practitioners, with some background in python and machine learning.  

Suggested reading

• AlRegib, G., & Prabhushankar, M. (2022). Explanatory Paradigms in Neural Networks: Towards relevant and contextual explanations. IEEE Signal Processing Magazine, 39(4), 59-72  
• Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision (pp. 618-626) 

• Prabhushankar, M., & AlRegib, G. (2021). Contrastive Reasoning in Neural Networks. arXiv preprint arXiv:2103.12329. 
• Goyal, Y., Wu, Z., Ernst, J., Batra, D., Parikh, D., & Lee, S. (2019, May). Counterfactual visual explanations. In International Conference on Machine Learning (pp. 2376-2384). PMLR 
• G. Kwon*, M. Prabhushankar*, D. Temel, and G. AlRegib, “Distorted Representation Space Characterization Through Backpropagated Gradients,” in IEEE International Conference on Image Processing (ICIP), Taipei, Taiwan, Sep. 2019. 
• G. Kwon, M. Prabhushankar, D. Temel, and G. AlRegib, “Backpropagated Gradient Representations for Anomaly Detection,” in Proceedings of the European Conference on Computer Vision (ECCV), SEC, Glasgow, Aug. 23-28 2020 
• Temel, D., Lee, J., & AlRegib, G. (2018, December). Cure-or: Challenging unreal and real environments for object recognition. In 2018 17th IEEE international conference on machine learning and applications (ICMLA) (pp. 137-144). IEEE 

• M. Prabhushankar, and G. AlRegib, “Introspective Learning : A Two-Stage Approach for Inference in Neural Networks,” in Advances in Neural Information Processing Systems (NeurIPS), New Orleans, LA,, Nov. 29 – Dec. 1, 2022, [PDF], [Code]. 

Speakers