Research

Automatic speech recognition (ASR) has been a grand challenge machine learning problem for decades. Our ongoing research in this area examines the use of deep learning models for distant and noisy recording conditions, multilingual, and low-resource scenarios.

Traffic is not just a nuisance for drivers: It’s also a public health hazard and bad news for the economy.

Clinical concept extraction (CCE) of named entities - such as problems, tests, and treatments - aids in forming an understanding of notes and provides a foundation for many downstream clinical decision-making tasks. Historically, this task has been posed as a standard named entity recognition (NER) sequence tagging problem, and solved with feature-based methods using hand-engineered domain knowledge. Recent advances, however, have demonstrated the efficacy of LSTM-based models for NER tasks, including CCE. This work presents CliNER 2.0, a simple-to-install, open-source tool for extracting concepts from clinical text. CliNER 2.0 uses a word- and character- level LSTM model, and achieves state-of-the-art performance. For ease of use, the tool also includes pre-trained models available for public use.

A system that can help you to design our own drone!

To further parallelize co-prime sampling based sparse sensing, we introduce Diophantine Equation in different algebraic structures to build generalized lattice arrays.
With strong relationship to generalized Chinese Remainder Theorem, the geometry properties in the remainder code space, a special lattice space, are explored.

Multi-robot path planning for robot swarms that can both fly and drive

EQ-Radio can infer a person’s emotions using wireless signals. It transmits an RF signal and analyzes its reflections off a person’s body to recognize his emotional state (happy, sad, etc.).

Data often has geometric structure which can enable better inference; this project aims to scale up geometry-aware techniques for use in machine learning settings with lots of data, so that this structure may be utilized in practice.

To enable privacy preservation in decentralized optimization, differential privacy is the most commonly used approach. However, under such scenario, the trade-off between accuracy (even efficiency) and privacy is inevitable. In this project, distributed numerical optimization scheme incorporated with lightweight cryptographic information sharing are explored. The affect on the convergence rate from network topology is considered.

We are developing a general framework that enforces privacy transparently enabling different kinds of machine learning to be developed that are automatically privacy preserving.