News

04 May 2024

I will be participating in the CIFAR workshop: World Models: Causality, Neuroscience, and AI Safety in Tübingen in the middle of June. I’m looking forward to writing up a summary of our discussions on the advancement of LLMs and their applications to cognitive (neuro)science and AI safety.

01 May 2024

I’m heading to ICLR 2024 to participate in the Re-Align Workshop, where I will present some recent work on measuring the alignment between biological and artificial neural networks. Here is the link to the paper (arxiv) and here is a link to the Github repo with a bit of a shorter outline of the findings and code.

November 2023

This academic semester at the University of Alberta, I have been co-instructing the class Machine Learning and the Brain along with Alona Fyshe. This class covers three main areas: (i) visual neuroscience & computer vision models (ii) neurolinguistics & NLP models and (iii) reinforcement learning. Recordings of the taught components of this course are available on my new YouTube channel (NeuroAlex) at here. If visiting towards the end of November 2023, things might still be getting adjusted/edited. I will begin recording data processing, analysis tutorials alongside paper reviews for those interested in the field of NeuroAI.

Github Repo for Machine Learning & the Brain.

June 2023

We introduce Identifying Shared Decodable Concepts in the Human Brain using Image-Language Foundation Models (arXiv link). An exciting project led by one of the team (Cory Efird) combining fMRI data (from NSD) and multimodal neural network representations (CLIP) in order to map out decodable visual concepts that are shared amongst participants at a finer-grained detail than what was possible for. We are currently developing an amended version and plan to release a fully interactive dataset viewer to explore these exciting results.

October 2022

• I presented a Three Minute Thesis (3MT) talk at Amii’s TechAid event in Edmonton, Canada, covering the work completed during my PhD
• I joined the University of Alberta / Amii as a postdoctoral research fellow in the lab of Dr Alona Fyshe, co-supervised by Martha White, to work on projects exploring how brain data can be best used to develop novel machine learning algorithms. I will study the properties underlying brain data that enable such enhancements, alongside methods to incorporate Reinforcement Learning into this interdisciplinary research area

September 2022

Completed PhD Viva at University of Birmingham, UK. Thesis committee members were Dr Katja Kornysheva (Department of Psychology, UoB) and Juntao Yu (School of Computer Science, University of Essex, UK)

August 2022

I gave a talk at the Artificial Intelligence Seminar at the Alberta Machine Intelligence Institute (Amii), covering previous work studying neural decoding of language responses using deep neural networks (Transformers) as well as a general framework for how I believe brain data can be used in developing the next generation of machine learning models.

Publications

2024

Cory Efird, Alex Murphy, Joel Zylberberg, Alona Fyshe. Contrastive Decoding of Visual Concepts in the Brain. In prep.

Sijie Ling, Alex Murphy, Alona Fyshe. Exploring Temporal Sensitivity in the Brain using Multi-Timescale Language Models: an EEG Decoding Study. Under review.

Alex Murphy, Joel Zylberberg, Alona Fyshe. Correcting Biased Centered Kernel Alignment Measures in Biological and Artificial Neural Networks. ICLR 2024, Re-Align Workshop.

2023

Cory Efird, Alex Murphy, Joel Zylberberg, Alona Fyshe. Identifying Shared Decodable Concepts in the Human Brain Using Image-Language Foundation Models. Preprint.

2022

Alex Murphy, Bernd Bohnet, Ryan McDonald, and Uta Noppeney. 2022. Decoding Part-of-Speech from Human EEG Signals. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2201–2210, Dublin, Ireland. Association for Computational Linguistics.

Alex Murphy. 2022. Decoding Linguistic Information from EEG Signals. Unpublished doctoral dissertation. University of Birmingham. (Submitted, link to follow)

2017

Alex Murphy. 2017. Machine Learning & fMRI: Applications to Neurolinguistics. Unpublished Master’s dissertation. University of Copenhagen.