24/04/2024 at 13.00
Your data is not i.i.d.: Learning from Data Streams
Recording
Bernhard Pfahringer
Recording
Your data is not i.i.d.: Learning from Data Streams
University of Waikato, New Zealand
Bernhard Pfahringer received his PhD degree from the University of Technology in Vienna, Austria, in 1995. He is a Professor with the Department of Computer Science, and a co-director for the AI Institute at the University of Waikato in New Zealand. His interests span a range of data mining and machine learning sub-fields, with a focus on streaming, randomization, complex data, and, inevitably over the last few years, Deep Learning as well.
I will first give a quick summary of Learning from Data Streams, and of Continual Learning, including some recent work on Online Continual Learning. I will give an overview of the TAIAO project, which stands for “Time-Evolving Data Science and Artificial Intelligence for Advanced Open Environmental Science”. Finally, I will quickly present the works of my current and recently finished PhD students, comprising the following topics: – Advanced Adaptive Classifier Methods for Data Streams – SO-KNL: Self-optimising K-Nearest Leaves Regression Ensembles – Anomaly Detection in Streaming Data – AutoML for Data Streams – Self-supervised Feature Extractor Training for Alzheimer’s Disease Classification – Feature Extractor Stacking for Cross-domain Few-shot Learning – ML Approaches for Malware Classification based on Hybrid Artefacts – Using LLMs to assess cybersecurity thread notes – Fake News detection in Urdu – Normalising Flows for Environmental Data – Fast Clustering using GPUs, 13:00
Bernhard Pfahringer
University of Waikato, New Zealand
Bernhard Pfahringer received his PhD degree from the University of Technology in Vienna, Austria, in 1995. He is a Professor with the Department of Computer Science, and a co-director for the AI Institute at the University of Waikato in New Zealand. His interests span a range of data mining and machine learning sub-fields, with a focus on streaming, randomization, complex data, and, inevitably over the last few years, Deep Learning as well.