Ml-reading-group March 2024

ml-reading-group@lists.saarland-informatics-campus.de

5 participants
3 discussions

Wednesday Reading Group
by Siu Lun Chau 20 Mar '24

20 Mar '24

Dear all, This week’s reading group is hosted by CISPA, and we will have Anurag Singh presenting the paper on "Invariant Risk Minimisation Game" (https://arxiv.org/abs/2002.04692). The reading group will be on Wednesday at 2:30 pm. As you may know, CISPA has recently moved offices to St.Ingbert so the organisers have proposed the following solution to keep the reading group running: Both udS and CISPA book a room for their members to meet and the presenter will present online in their respective location. The link to join virtually can be found here: https://docs.google.com/spreadsheets/d/1vtgEezBqS4d_ACPt-emK2NT52x7nofX9jxg… Thank you, Alan

2 2

Updated paper for today's RG
by Pablo 06 Mar '24

06 Mar '24

Hello All, I've decided to change the paper for today's reading group. The proposed one had too much text and it would not be an "enjoyable” experience for the RG! This is the paper I'll be presenting: https://arxiv.org/abs/2302.02941 Sorry for the last-minute change, but I think it's for the best! See you later Best, Pablo

2 4

Wednesday reading group
by Adrián Javaloy 05 Mar '24

05 Mar '24

Dear all, This week's reading group is hosted by *udS*, and we will have *Pablo Sanchez* talking about the quite recent paper Statistics without Interpretation: A Sober Look at Explainable Machine Learning <https://arxiv.org/abs/2402.02870>. You can find an abstract of the paper below. As always, the reading group will be on *Wednesday at 2:30pm at room 0.01 of the E1.7 building*. I am in the process of find a free room with a working projector, so I will update you if the room changes at some point, for now we meet where always. If someone plans to join remotely, please *let me know in advance.* *Abstract* In the rapidly growing literature on explanation algorithms, it often remains unclear what precisely these algorithms are for and how they should be used. We argue that this is because explanation algorithms are often mathematically complex but don't admit a clear interpretation. Unfortunately, complex statistical methods that don't have a clear interpretation are bound to lead to errors in interpretation, a fact that has become increasingly apparent in the literature. In order to move forward, papers on explanation algorithms should make clear how precisely the output of the algorithms should be interpreted. They should also clarify what questions about the function can and cannot be answered given the explanations. Our argument is based on the distinction between statistics and their interpretation. It also relies on parallels between explainable machine learning and applied statistics. Cheers, Adrián

2 1

2025

2024

2023

Ml-reading-group March 2024

2025

2024

2023

Ml-reading-group March 2024 ----- 2025 ----- February 2025 January 2025 ----- 2024 ----- December 2024 November 2024 October 2024 September 2024 August 2024 July 2024 June 2024 May 2024 April 2024 March 2024 February 2024 January 2024 ----- 2023 ----- December 2023 November 2023 October 2023

Ml-reading-group March 2024