Neural Architectures Co-evolved for Resource Efficiency

We grow neural architectures instead of hand-designing them: evolutionary search discovers compact recurrent models for time series forecasting, small enough to run on the device where the data is generated, yet competitive with architectures many times their size.

What we work on

Research directions

Four threads, each feeding the next: better search produces better forecasters, forecasters find problems worth solving in other disciplines, and the whole thing is taught forward to students who have never written a line of machine learning code.

Evolutionary Neural Architecture Search

We develop neuroevolution methods that automatically design recurrent neural networks, with a focus on discovering compact architectures that rival much larger hand-designed models. The emphasis is on ultra-lightweight designs suitable for edge deployment, where memory, power, and latency constraints rule out conventional deep learning pipelines.

neuroevolution RNN design edge AI

EXAMM Neural Architecture

Time Series Forecasting

We work on forecasting methods for multivariate time series across both offline and online settings. In the offline regime, the focus is on designing accurate, compact models trained on historical data; in the online regime, we develop methods that adapt continuously as data distributions shift over time. Applications span industrial sensor data (coal-fired power plant operations) and financial forecasting (stock return prediction and portfolio trading).

multivariate time series offline and online forecasting non-stationary data

ONE-NAS: Online NeuroEvolution for time series

Cross-Disciplinary Edge AI Applications

A growing line of our work applies neuroevolved lightweight models to domains outside traditional ML benchmarks, partnering with researchers in engineering, energy, and finance. The goal is forecasting and decision-support systems that run on the device where the data is generated.

applied ML interdisciplinary collaboration resource-constrained inference

Example of an EXAMM generated RNN architecture

An evolved RNN architecture produced by EXAMM

AI Education & Pedagogy

We research how to broaden AI literacy beyond computer science majors. The lab runs the Kean Cup AI Competition as a venue for students across disciplines to work on applied AI projects, and leads an AI research stream in the Group Summer Scholars Research Program for high school students. This line of work explores curriculum design, assessment, and the pedagogy of AI for non-specialists.

AI literacy interdisciplinary education outreach

Department of Computer Science and Technology, Kean University, Union, New Jersey.