Learning, deciding, and the ways both go wrong — including the alignment problem, which I think is mostly an institutional problem wearing a technical hat.

Beyond Earning to Give October 10, 2025
The EA Road to AI

The Bayesian Computation Wishlist July 30, 2025
The fundamental operations that define the practice of Bayesian inference.

A tutorial on Bayesian Belief Bellman equations July 29, 2025
And Information Directed Sampling

Global coordination problems via game theory July 24, 2025
Why AGI is not like climate change

The DAO Proving Ground July 9, 2025
An Engineering Approach to Building Aligned Organizations

Committed to Fidelity July 3, 2025
A Unified Model for Multi-Fidelity Bandits

Why are future rewards worth less than present rewards? June 17, 2025
A Bayesian Derivation of Discounting

Goal Hijacking June 16, 2025
How Intrinsic Motivation Can Derail Goal-Directed Behavior

The Hammer and the Nuke June 16, 2025
Why AI Robustness is a Necessary Condition for AI Alignment

Turing versus Scherbius June 1, 2025
The card game

The Alignment Problem May 23, 2025
Future AI Overlords vs. Present Corporate Greed

Affirmative Action's Simplistic Approach May 20, 2025
Using a 1-Dimensional Solution for a Multi-Dimensional Problem

Projection into the typical set: PITS August 10, 2024
A new approach to solving inverse problems

Diffusion posterior sampling August 10, 2024
A review of recent work

Score functions, denoising and diffusion August 10, 2024
Diffusion is just stacked denoising score matching!

The behaviour of neural flows August 1, 2024
Neural nets can struggle to learn very simple flows.

Language models are all you need November 10, 2022
for SMILES-based chemistry

Requests for research April 10, 2020
Some ideas from my masters.

The fable of the caterpillar February 25, 2019
A fun intro to my masters topic; abstraction for efficient reinforcment learning.

The advantages of backward reasoning. October 20, 2018
A simple exploration of what can be gained by reasoning backwards from your goal.

Real time bandits September 5, 2018
Rewarding a twitter bot is more complicated than I imagined.

Automated science August 22, 2018
Model-based reinforcement learning, symbolic AI, and the limits of efficient learning.

ACL2018 July 25, 2018
Notes and thoughts from attending ACL.

Unsupervised skip connections May 30, 2018
How can we use ladder nets in the unsupervised setting?

Covering letter for my first PhD proposal May 14, 2018
Principles of neural design, automated science, general reinforcement learning

Tax optimisation as an adversarial game March 27, 2018
A new perspective on tax law

Deep learning September 20, 2017
An intro to deep learining for COMP421.

Saddles, Splitting, and Reparameterization September 26, 2016
Dynamically Growing Neural Networks

A functional type of non-linear algebra? September 26, 2016
Functional programming plus neural network architectures.