Research

Conference

	How Robust are Estimated Effects of Nonpharmaceutical Interventions against COVID-19? (2020), NeurIPS Spotlight paper, 4th author / 10. Video.
	Safety Properties of Inductive Logic Programming (2020), AAAI SafeAI workshop, 1st author / 3. Video.

Journal

Inferring the effectiveness of government interventions against COVID-19 (2020), Science,
8th author / 19. Explainer, discussion, app.

1

Preprint

	Understanding the effectiveness of interventions in Europe's second wave of COVID-19 (2021), 4th author / 22. Explainer.
	Legally Grounded Fairness Objectives (2020), 2nd author / 3. Blogpost.

Software

ProlexaPlus (2020): Bringing modern language modelling into Prolog for some reason.
Py2HTK (2017): Python wrapper for the Hidden Markov ToolKit.

Popular

The academic contribution to AI safety seems large (2020)
Existential risk as common cause (2018)
Side effects in Gridworlds (2018). Developed further.

Teaching

2019: TA for the fearsome COMS30007: Bayesian Machine Learning
2020: Lead TA for COMS20010: Algorithms 2.

Read if: you want to know what government moves worked in the first wave of the pandemic. You want to see how sensitive results are. You want to see what computer scientists can do when they attack en masse.

Credit to James Walsh for the academic SVGs.