Ban Kawas is a member of the Scalable Knowledge Intelligence (SKI) group working on (1) Auto AI for Text, (2) the automatic generation of Information Extraction Rules from user-specified examples, (3) Explainable AI (XAI), and (4) Text Understanding with Human-in-the-loop.
Ban's research spans several areas of AI, Machine Learning, and Decision-Making & Planning Under-Uncertainty; including Deep Reinfrocement Learning, Semi-supervised Learning, Feature Selection, Representation Learning & Embeddings, Dimensionality Reduction & Feature Extraction methods, Learning from Small-Data, Generative Models & Data Augmentation, Bayesian Optimization, Robust Optimization & Adversarial Learning, and Anomaly & Outlier Detection algorithms. Ban is also interested in Recommender Systems, Ranking Models, Automatic Labeling, and Knowledge Extraction.
Previously, Ban was the AI Scientific Lead in the Science to Solutions Group & the Technical Lead for the Consortium for Sequencing the Food Supply Chain. Ban continues to leverage her expertise in Science to Solutions projects; mainly to (1) Analyze genomics and metagenomics data for the development of novel representations of biological samples, and the design of Outlier/Anomaly Detection Systems for Food-Safety, and for (2) Salient Features Characterization and Extraction for High-Dimensional Low-Sample Size (HDLSS) Data for the Accelerated Discovery of Polymer & Material Design.