In the past decade, Machine Learning has become one of the major opportunities to perform breakthrough cross-domain research in conventional scientific fields, through ever increasing hardware performance and abundant availability of data through ubiquitous connectivity. The embedded Signal Processing and Machine Learning (eSPML) Lab is chartered to research with two main focuses:
- data-driven machine learning algorithms to complement and/or replace model-based advanced signal processing algorithms and
- related novel embedded implementations through applying digital, mixed signal and/or analog integrated circuit design in new ways.
Machine Learning, Neural Networks and Deep Learning approaches are meanwhile becoming increasingly prominent in various computer science and circuits and systems societies, where system identification and dimensionality reduction are the challenge. The underlying idea of this lab is to bring diverse, yet complementary skillsets in these fields together to jointly achieve solution inspiring research results.