Calum Heggan
Currently researching within the fields of Few-Shot Machine Learning and Self-Supervised Representation Learning , I am a 4th year PhD candidate at the University of Edinburgh working with both the IPAB and IDCOM groups. I work alongside Mehrdad Yaghoobi (Singal Processing), Timothy Hospedales (Informatics) and Sam Budgett (Thales UK).
I am currently looking for jobs and open to opportunities. Please get in contact if you would like to discuss.
Alongside the mentioned topics, my research interests also include meta-learning and anomaly detection, and how they apply to a variety of downstream tasks and domains.
Conferences “From Pixels to Waveforms: Evaluating Pre-trained Image Models for Few-Shot Audio Classification” accepted at IJCNN24. Paper and code coming soon.
Other Started writing Thesis
Conferences “On the Transferability of Large-Scale Self-Supervision to Few-Shot Audio Classification” is accepted to ICASSP SASB. Code to be released soon, and formally presented in Seoul April 2024
Conference Presentations MT-SLVR presented at INTERSPEECH 23
Publications MT-SLVR and its relevant code are officially released
Conferences MT-SLVR (A new multi-task representation learning approach) is accepted to INTERSPEECH23. To be released soon, and formally presented in August 2023
Workshops Co-hosted an international Few-Shot Learning workshop for Thales UK’s “FoundAItion” (5/5/23)
Conference Presentations MetaAudio presented at ICANN22 Session 2 (6/9/22)
Conferences MetaAudio accepted to ICANN22. To be presented in early September
Publications MetaAudio, the first few-shot audio classification benchmark is now live on arXiv and submitted to ICANN22