Jun 20, 2024 
Another year, another update: I have published two blog posts on the ICLR 2024 blog post trackâwhile working at Midjourney. Iâm grateful for the opportunity to work on this open research on the side.
In particular, one of the blog posts was selected as a Highlight of the blog post track:
Both blog posts are available on the ICLR 2024 Blogpost Track website.

Dec 1, 2023 
Another year, another set of papers. This year was dominated by writing up my thesis and defending it. Iâm also very happy to have joined MidJourney as a Research Scientist at the end of September. Thus, this year was mostly about wrapping up some loose ends into papers:
 âDoes âDeep Learning on a Data Dietâ reproduce? Overall yes, but GraNd at Initialization does notâ (Kirsch, 2023)
 âBlackBox Batch Active Learning for Regressionâ (Kirsch, 2023)
 âStochastic Batch Acquisition: A Simple Baseline for Deep Active Learningâ (Kirsch et al., 2023)
And finally my thesis: âAdvancing Deep Active Learning & Data Subset Selection: Unifying Principles with InformationTheory Intuitionsâ (Kirsch, 2023)

Dec 1, 2022 
Very happy to have published a few papers at TMLR (and coauthored one presented at ICML) this year and to have (joint) coauthored papers that we will present at CVPR and AISTATS next year:
 âPrioritized Training on Points that are Learnable, Worth Learning, and not yet Learntâ, ICML 2022 (Mindermann* et al., 2022)
 âA Note onâAssessing Generalization of {SGD} via Disagreement"", TMLR(Kirsch & Gal, 2022)
 âUnifying Approaches in Active Learning and Active Sampling via Fisher Information and InformationTheoretic Quantitiesâ, TMLR (Kirsch & Gal, 2022)
 âDeterministic Neural Networks with Appropriate Inductive Biases Capture Epistemic and Aleatoric Uncertaintyâ, CVPR 2023 (Highlight) (Mukhoti* et al., 2023)
 âPredictionOriented Bayesian Active Learningâ, AISTATS 2024 (Bickford Smith* et al., 2023)
Several of these can be traced to workshop papers, which we were able to expand and polish into full papers.

Jul 24, 2021 
Seven workshop papers at ICML 2021 (out of which five are first author submissions):
Two papers and posters at the Uncertainty & Robustness in Deep Learning workshop:
Four papers (posters, one spotlight) at the SubSetML: Subset Selection in Machine Learning: From Theory to Practice workshop:
Active Learning under Pool Set Distribution Shift and Noisy Data
Andreas Kirsch, Tom Rainforth, Yarin Gal
SPOTLIGHT
(also accepted at the Human in the Loop Learning (HILL) workshop)
or as Poster PDF version as download.
A Simple Baseline for Batch Active Learning with Stochastic Acquisition Functions
Andreas Kirsch, Sebastian Farquhar, Yarin Gal
(also accepted at the Human in the Loop Learning (HILL) workshop)
or as Poster PDF version as download.
A Practical & Unified Notation for InformationTheoretic Quantities in ML
Andreas Kirsch, Yarin Gal
or as Poster PDF version as download.
Prioritized training on points that are learnable, worth learning, and not yet learned
SĂ¶ren Mindermann, Muhammed Razzak, Winnie Xu, Andreas Kirsch, Mrinank Sharma, Adrien Morisot, Aidan N. Gomez, Sebastian Farquhar, Jan Brauner, Yarin Gal
or as Poster PNG version as download.
One paper (poster) at the Neglected Assumptions In Causal Inference workshop:

Feb 23, 2021 
Lecture on âBayesian Deep Learning, Information Theory and Active Learningâ for Oxford Global Exchanges. You can download the slides here.

Feb 21, 2021 
Deterministic Neural Networks with Appropriate Inductive Biases Capture Epistemic and Aleatoric Uncertainty has been uploaded to arXiv as preprint. Joint work with Jishnu Mukhoti, and together with Joost van Amersfoort, Philip H.S. Torr, Yarin Gal. We show that a single softmax neural net with minimal changes can beat the uncertainty predictions of Deep Ensembles and other more complex singleforwardpass uncertainty approaches.

Dec 10, 2020 
Unpacking Information Bottlenecks: Unifying InformationTheoretic Objectives in Deep Learning was also presented as a poster at the âNeurIPS Europe meetup on Bayesian Deep Learningâ.
You can find the poster below (click to open):
or as PDF version to download.

Jul 17, 2020 
Two workshop papers have been accepted to Uncertainty & Robustness in Deep Learning Workshop at ICML 2020:
 Scalable Training with Information Bottleneck Objectives, and
 Learning CIFAR10 with a Simple Entropy Estimator Using Information Bottleneck Objectives
both together with Clare Lyle and Yarin Gal. They are based on Unpacking Information Bottlenecks: Unifying InformationTheoretic Objectives in Deep Learning for the former, and an application of the UIB framework for the latter: we can use it to train models that perform well on CIFAR10 without using a crossentropy loss at all.

Mar 27, 2020 
Unpacking Information Bottlenecks: Unifying InformationTheoretic Objectives in Deep Learning, together with Clare Lyle and Yarin Gal, has been uploaded as preprint to arXiv. It examines and unifies different Information Bottleneck objectives and shows that we can introduce simple yet effective surrogate objectives without complex derivations.

Sep 4, 2019 
BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learning got accepted into NeurIPS 2019. See you all in Vancouver!
