Andy King

@AtoAndyKing
0 Followers
52 Following
20 Posts
King's College London computer scientist, research in #MachineLearning & #AI/#MedicalAI for #MedicalImaging, teacher, author, father of 2, former camel owner, imposter
Group www:http://kclmmag.org
KCL www:https://www.kcl.ac.uk/people/andrew-king
Orcidhttps://orcid.org/0000-0002-9965-7015
I enjoyed this paper, "The Problem with Metrics is a Fundamental Problem for AI". In ML, often we glamorize algorithmic or model improvements and ignore the importance of what is around that such as the data used, metrics for success, or a good UI for the humans trying to get something done with our models. It is important that ML systems optimize for the right thing, and they often do not because metrics for success are picked too sloppily and hastily.
https://arxiv.org/abs/2002.08512
The Problem with Metrics is a Fundamental Problem for AI

Optimizing a given metric is a central aspect of most current AI approaches, yet overemphasizing metrics leads to manipulation, gaming, a myopic focus on short-term goals, and other unexpected negative consequences. This poses a fundamental contradiction for AI development. Through a series of real-world case studies, we look at various aspects of where metrics go wrong in practice and aspects of how our online environment and current business practices are exacerbating these failures. Finally, we propose a framework towards mitigating the harms caused by overemphasis of metrics within AI by: (1) using a slate of metrics to get a fuller and more nuanced picture, (2) combining metrics with qualitative accounts, and (3) involving a range of stakeholders, including those who will be most impacted.

arXiv.org
Papers accepted at #Neurips are now visible in OpenReview https://openreview.net/group?id=NeurIPS.cc/2022/Conference
NeurIPS 2022 Conference

Welcome to the OpenReview homepage for NeurIPS 2022 Conference

OpenReview

Hello, New World!

I’m Professor in Computational Imaging and AI in Medicine at TU Munich, Director of the Helmholtz Munich Institute of Machine Learning in Biomedical Imaging, and Chair in Computational Imaging at King’s College London.

Looking forward to some great scientific discourse in those fields.

#introduction

Tip for newbies: filling in the text description for pics posted is, I have learned, really appreciated. There’s a sizeable community of the visually impaired over here and doing it means they’re not excluded. Also, it’s fun to do. If you’re on your phone, tap on the pic and start typing. (I didn’t realise I could go that) On the desktop app you’re given a prompt. It’s super easy.
If you want to delete all your old Twitter DMs (which I think is a good idea even if Twitter might retain copies of them somewhere), and also your historical likes, RTs, and any tweets you don’t want to keep, I made a Twitter app for that.
It’s called Semiphemeral and it’s pretty easy to use. I posted instructions for how to do it https://twitter.com/semiphemeral/status/1591258781098594304?s=46&t=L4aWvmyNxEFRL6eULgeO-A
Semiphemeral on Twitter

“As you watch in horror/delight as Elon burns this site to the ground you might be pondering your privacy 🤔. Do you have YEARS of tweets, likes, and DMs? https://t.co/EF6AVCPm2O Gather 'round, friends, while I show you how to DELETE THEM ALL (or as much as Twitter's API allows)”

Twitter

#PhD students of #Mastodon, I have created a Mastadon group for sharing PhD-related posts using @phdstudents when posting a PhD student toot. Let's help the group to grow and share our PhD student experiences.

#Students #postgraduate #research #connectivity #PhDStudents

#introduction

I am computer science professor in machine learning for imaging at Imperial College London. Primarily working on medical imaging applications and interested in anything related to ML robustness, fairness, causality.

What is Mastodon?
What are the pros and cons of Mastodon over Twitter?
Should I make the switch?

All that and more in our latest blog!

#academia #PhD #science #xp @OpenAcademics @AcademicChatter

Read it here:
https://thestrugglingscientists.com/mastodon-for-academics/

The Complete Guide to Mastodon and Twitter for Academics - The Struggling Scientists 2025

In this article, we will explore the differences between Mastodon and Twitter, as well as the benefits of using Mastodon for academics.

The Struggling Scientists

Time for a quick #Introduction

I am a Research Scientist at Spotify's Advanced Causal Inference Lab. I recently got my PhD from Imperial College London working with @BernhardKainz

My main research focus is causal inference in ML, but I also enjoy reading about and working in computer vision, healthcare related ML and space exploration !

Have just realised that #academics can use @ORCID_Org to
#verify themselves.

Just include your full mastodon link (like this: https://mas.to/@marekmcgann) in the "Websites and Social Links" section in ORCID, then include link to your ORCID record in your Mastodon profile.

Hurray for distributed digital identities!