01/2026: Notes on data parallelism in training large language models - “How many GPUs do you need to train a transformer?” [notes]

01/2026: Notes on policy gradient methods. [notes]

09/2025: Talk titled “Orthogonality in neural network” for Scaleout Edges. [slides, notes]

09/2025: Talk about flow matching and diffusion models at our group seminar. [notes, code]

06/2025: Half-time seminar titled “Learning from distributed and heterogeneous data” for Department of Information Technology, Uppsala University. [slides]

06/2025: Talk about distributed and federated learning for master course Distributed and Parallel Computing as a guest lecture. [slides]

10/2024: Talk about Bayesian neural network at our group seminar. [notes]

10/2024: Poster presentation titled “Is logit adjustment a free lunch for heterogeneous federated learning?” at eSSENCE 2024. [poster]

05/2024: Talk titled “Federated learning for predicting compound mechanism of action based on image-data from cell painting” for AI4Research, Uppsala University. [slide]

04/2024: Talk introducing denoising diffusion probabilistic model (DDPM) for PhD course Bayesian Inference. [slide, cheat-sheet for Bayesian Inference]

12/2023: Talk introducing federated learning from the perspective of optimization for Centre for Interdisciplinary Mathematics, Uppsala University. [slide]