Public repo for The Alan Turing Institute's reading group on foundation models as part of the activities of Foundational AI theme.
If you're based at the Turing, follow #robots-in-disguise on the Turing Slack for the most recent updates.
To see all the slides and reading materials for previous sessions, see the archive.
Note that this originated from the Research Engineering Team's reading group on Transformers.
The group meets every week on Mondays at 11-12. Everyone is welcome to join! If you have any questions email Ryan or Fede and remember to go through our Code of Conduct before joining.
Please get in touch if you would like to give a talk (either about your research or a topic you think is relevant to the reading group) add suggestions and emoji preferences to the list of proposed topics on HackMD!
| Date | Topic | Room | Lead |
|---|---|---|---|
| 20/05/24 | Technical: KAN: Kolmogorov-Arnold Networks | Ursula Franklin | Andrew Duncan |
| 04/06/24 | Invited Talk: Are we ready for attacks on machine learning? | Enigma (2.30pm) | Nicholas Carlini |
| 10/06/24 | REG Hack Week 👋 | David Blackwell | |
| 17/06/24 | Research at Turing: Edge AI | David Blackwell | Liam Fletcher, Kat Goldmann, Colin Laganier and others |
| 24/06/24 | Invited Talk: Improving LLMs for low-resource African languages | Ursula Franklin (3pm) | Ed Bayes and others |
| 01/07/24 | Technical: A perspective on the fundamentals of transformers | Ursula Franklin | Ed Gunn |
| 08/07/24 | TBC: Conference Overview: Coling/LREC | Cipher | Fede Nanni |
| 15/07/24 | Invited Talk: [TBC] | David Blackwell | Gavin Abercrombie |
| 22/07/24 | Invited Talk: Designing a Value-driven GAI Framework for Social Good: Embedding Social Good Values into GAI Models | Ursula Franklin | Victor OK Li, Jacqueline CK Lam and Jon Crowcroft |
| 29/07/24 | TBC | Ursula Franklin | TBC |
| 05/08/24 | TBC | Ursula Franklin | TBC |
| 12/08/24 | TBC | Ursula Franklin | TBC |
| 19/08/24 | TBC | David Blackwell | TBC |
| 26/08/24 | TBC | David Blackwell | TBC |
| 02/09/24 | TBC | David Blackwell | TBC |
| 09/09/24 | TBC | Ursula Franklin | TBC |
| 16/09/24 | TBC | David Blackwell | TBC |
| 23/09/24 | TBC | David Blackwell | TBC |
Main
Abstract It has now been a decade since the first adversarial examples were demonstrated on deep learning models. And yet, even still, we can not robustly classify MNIST images better than LeNet-5 or ImageNet images better than AlexNet. But now, more than ever, we need robust machine learning models. And not only robust to evasion attack: but also robust to poisoning, stealing, and many other attacks. In this talk I survey the current progress we have made on adversarial machine learning. While we have made many significant advances in making attacks practical, we have had made considerably less progress on defences. Making progress towards addressing these challenges will be of the highest importance in the coming years.