Beyond Earning to Give
The EA Road to AI
The Bayesian Computation Wishlist
The fundamental operations that define the practice of Bayesian inference.
A tutorial on Bayesian Belief Bellman equations
And Information Directed Sampling
Global coordination problems via game theory
Why AGI is not like climate change
The DAO Proving Ground
An Engineering Approach to Building Aligned Organizations
Committed to Fidelity
A Unified Model for Multi-Fidelity Bandits
Why are future rewards worth less than present rewards?
A Bayesian Derivation of Discounting
Goal Hijacking
How Intrinsic Motivation Can Derail Goal-Directed Behavior
The Hammer and the Nuke
Why AI Robustness is a Necessary Condition for AI Alignment
Turing versus Scherbius
The card game
The Alignment Problem
Future AI Overlords vs. Present Corporate Greed
Affirmative Action's Simplistic Approach
Using a 1-Dimensional Solution for a Multi-Dimensional Problem
Projection into the typical set: PITS
A new approach to solving inverse problems
Diffusion posterior sampling
A review of recent work
Score functions, denoising and diffusion
Diffusion is just stacked denoising score matching!
The behaviour of neural flows
Neural nets can struggle to learn very simple flows.
Language models are all you need
for SMILES-based chemistry
Requests for research
Some ideas from my masters.
The fable of the caterpillar
A fun intro to my masters topic; abstraction for efficient reinforcment learning.
The advantages of backward reasoning.
A simple exploration of what can be gained by reasoning backwards from your goal.
Real time bandits
Rewarding a twitter bot is more complicated than I imagined.
Automated science
Model-based reinforcement learning, symbolic AI, and the limits of efficient learning.
ACL2018
Notes and thoughts from attending ACL.
Unsupervised skip connections
How can we use ladder nets in the unsupervised setting?
Covering letter for my first PhD proposal
Principles of neural design, automated science, general reinforcement learning
Tax optimisation as an adversarial game
A new perspective on tax law
Deep learning
An intro to deep learining for COMP421.
Saddles, Splitting, and Reparameterization
Dynamically Growing Neural Networks
A functional type of non-linear algebra?
Functional programming plus neural network architectures.