[ML] Edge Machine Learning for Resource-constrained IoT Devices

In this project, we propose an edge computing solution that enables distributed machine learning on resource constrained IoT devices. We develop a scalable algorithm to automatically dispatch neural networks to edge devices. The design is tested on Raspberry Pis for image classification tasks.

[ML-Theory] On the Intensity Estimation of Poisson Process

Poisson process plays a fundamental role in the theory and application of stochastic process. However, the intensity estimation of inhomogeneous Poisson process remains difficult. In this project, we compare the performance of three existing estimation methods: smoothing kernel method, Mercer kernel method, and Bayesian method.

[ML] Pattern Recognition for Brain Waves

Brain waves have been widely used to measure human’s physical and mental activities. With the help of cutting edge machine learning technologies, we successfully detect two activity patterns from EEG signal: blink and concentration.