- This event has passed.
Talk On Making Machine Learning Models Safer and Better: Data and Model Perspectives
April 12 @ 4:00 PM - 5:30 PM IST
Abstract: As machine learning systems are increasingly deployed in real-world settings like healthcare, finance, and scientific applications, ensuring their safety and reliability is crucial. However, many state-of-the-art ML models still suffer from issues like poor out-of-distribution generalization, sensitivity to input corruptions, requiring large amounts of data, and inadequate calibration – limiting their robustness and trustworthiness for critical real-world applications.
In this talk, I will present a broad overview of different safety considerations for modern ML systems. I will then proceed to discuss our recent efforts in making ML models safer from two complementary perspectives – (i) manipulating data and (ii) enriching the model capabilities by developing novel training mechanisms. First, I will discuss our work on designing new data augmentation techniques for object detection followed by demonstrating how, in the absence of data from desired target domains of interest, one could leverage pre-trained generative models for efficient synthetic data generation. Next, I will introduce a new paradigm of training deep networks called model anchoring and show how one could achieve similar properties to an ensemble but through a single model. I will specifically discuss how model anchoring can significantly enrich the class of hypothesis functions being sampled and demonstrate its effectiveness through its improved performance on several safety benchmarks. Finally, I will present our efforts in proactively identifying samples on which a model would fail through a novel model counterfactual synthesis technique by leveraging foundation models (e.g., GPT family and CLIP). I will then conclude by highlighting exciting future research directions for producing robust ML models through leveraging multi-modal foundation models.