Fundamental machine learning topics:

Topics we are currently exploring:

1. Quantifying the influence of individual datapoints on the output of neural network models Papers
   • Leave-One-Out Distinguishability in Machine Learning with Jiayuan Ye, Reza Shokri and Soufiane Hayou
   • Quantifying and Localizing Usable Information Leakage from Neural Network Gradients with Fan Mo, Mohammad Malekzadeh, Soteris Demetriou, Deniz Gunduz and Hamed Haddadi

2. Developing new optimisation schemes for spiking neural networks that avoid the use of gradients.

3. Understand the relevant information in the data through dataset reconstruction.

4. Improve task generalisation by modifying the dataset used and the way we represent the data. Papers
   • Efficient regression with deep neural networks: how many datapoints do we need? with Daniel Lengyel

Computational neuroscience topics:

Applications we are currently working on:

1. Finding neural latent embeddings that can reveal underlying brain mechanisms during diverse behavior.

2. Quantifying similarities between artificial neural networks and the brain to shed light on mechanisms that are most like those used by the brain.

3. Finding biomarkers to improve deep brain stimulation-based treatment for Parkinson’s disease. Papers

   • Explainable Deep Learning for Arm Classification During Deep Brain Stimulation - Towards Digital Biomarkers for Closed-Loop Stimulation with Mathias Haugland, Yen Tai and Shlomi Haar

---

This is a jekyll based resume template made by Ankit Sultana