Imaging Science Seminar: Dimah Dera

Imaging Science Seminar
Towards Machine Self-Awareness – A Bayesian Framework for Uncertainty Propagation in Deep Neural Networks
Dr. Dimah Dera
Endowed Assistant Professor, Electrical and Computer Engineering
University of Texas Rio Grande Valley

This seminar will introduce new deep learning methods that are able to quantify their uncertainty in the decision and self-assess their performance, are robust to adversarial attacks and can even expose an attack from ambient noise. The main contribution of this work is establishing the theoretical and algorithmic foundations of uncertainty or belief propagation through complex deep learning models by adopting powerful frameworks from density tracking in non-linear and non-Gaussian dynamical systems.

Abstract:
Deep neural networks (DNNs) have surpassed human-level accuracy in various fields, including object identification, face recognition, voice recognition and image segmentation. Deep learning techniques hold the promise of emerging technologies, such as self-driving cars and autonomous unmanned aircraft systems, smart cities infrastructure, personalized treatment in medicine, and cybersecurity. However, unlike Humans who have a natural cognitive intuition for probabilities, DNN systems, being inherently deterministic, are unable to evaluate their confidence in the decisions. To truly deserve its name, an artificial intelligence system must be aware of its limitations and have the capacity for insightful introspection. This seminar will introduce new deep learning methods that are able to quantify their uncertainty in the decision and self-assess their performance, are robust to adversarial attacks and can even expose an attack from ambient noise. The main contribution of this work is establishing the theoretical and algorithmic foundations of uncertainty or belief propagation through complex deep learning models by adopting powerful frameworks from density tracking in non-linear and non-Gaussian dynamical systems. Bayesian inference provides a principled approach to reason about model confidence or uncertainty by estimating the posterior distribution of the unknown parameters given the observed data. The challenge in DNNs is the multi-layer stages of non-linearities in deep learning models, which makes propagation of high-dimensional distributions mathematically intractable. Drawing upon powerful statistical frameworks for density propagation in non-linear and non-Gaussian dynamical systems, we introduce Tensor Normal distributions as priors over the network parameters and derive a first-order Taylor series mean-covariance propagation framework. We subsequently extend this first-order approximation to an unscented framework that propagates sigma points through the model layers. The unscented framework is shown to be accurate to at least the second-order approximation of the posterior distribution. The uncertainty in the output decision is given by the propagated covariance of the predictive distribution. Experimental results on benchmark datasets, including MNIST, CIFAR-10, real-world synthetic aperture radar (SAR) and Brain tumor segmentation (BraTS 2015), demonstrate: 1) superior robustness against Gaussian noise and adversarial attacks; 2) self-assessment through predictive confidence that decreases quickly with increasing levels of ambient noise or attack; and 3) an ability to detect a targeted attack from ambient noise.

Speaker Bio:
Dimah Dera (Member IEEE) is an Endowed Assistant Professor in the Department of Electrical and Computer Engineering at the University of Texas Rio Grande Valley. She received her Ph.D. in Electrical and Computer Engineering from Rowan University. Dimah received the National Science Foundation (NSF) Computer and Information Science and Engineering Research Initiation Initiative (CRII) and NSF Research Experiences for Undergraduates (REU) supplement awards in 2022 for her current research focusing on robust machine learning and time-series analysis. She also received the Fred W. and Frances H. Rusteberg Endowment Fellowship award in 2022. She won the Best Student Paper Award at the 2019 IEEE International Workshop on Machine Learning for Signal Processing (MLSP’19) and the Runner-up Best Paper Award at the 2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM’15). She is the recipient of the NJ Health Foundation Research Grant award (2021), the IEEE Philadelphia Sections Benjamin Franklin Key award (2021), the STEM Innovator to Watch award from the NJ Tech Council (2018) and the NSF iREDEFINE Professional Development award (2017). Her research interests are robust machine learning, focusing on Bayesian deep learning, statistical tracking, and optimization.

Intended Audience:
Undergraduates, graduates, and experts. Those with interest in the topic.