collaborators

6 papers

cs.AI2024

On the consistent reasoning paradox of intelligence and optimal trust in AI: The power of 'I don't know'

Alexander Bastounis, Paolo Campodonico, Mihaela van der Schaar +2

We introduce the Consistent Reasoning Paradox (CRP). Consistent reasoning, which lies at the core of human intelligence, is the ability to handle tasks that are equivalent, yet des…

cs.AI2024

Stealth edits to large language models

Oliver J. Sutton, Qinghua Zhou, Wei Wang +4

We reveal the theoretical foundations of techniques for editing large language models, and present new methods which can do so without requiring retraining. Our theoretical insight…

math.OC2023

When can you trust feature selection? -- II: On the effects of random data on condition in statistics and optimisation

Alexander Bastounis, Felipe Cucker, Anders C. Hansen

In Part I, we defined a LASSO condition number and developed an algorithm -- for computing support sets (feature selection) of the LASSO minimisation problem -- that runs in polyno…

math.OC2023

When can you trust feature selection? -- I: A condition-based analysis of LASSO and generalised hardness of approximation

Alexander Bastounis, Felipe Cucker, Anders C. Hansen

The arrival of AI techniques in computations, with the potential for hallucinations and non-robustness, has made trustworthiness of algorithms a focal point. However, trustworthine…

cs.LG2023

The Boundaries of Verifiable Accuracy, Robustness, and Generalisation in Deep Learning

Alexander Bastounis, Alexander N. Gorban, Anders C. Hansen +5

In this work, we assess the theoretical limitations of determining guaranteed stability and accuracy of neural networks in classification tasks. We consider classical distribution-…

cs.LG2023

How adversarial attacks can disrupt seemingly stable accurate classifiers

Oliver J. Sutton, Qinghua Zhou, Ivan Y. Tyukin +3

Adversarial attacks dramatically change the output of an otherwise accurate learning system using a seemingly inconsequential modification to a piece of input data. Paradoxically,…