The Dan MacKinlay stable of variably-well-consider’d enterprises danmackinlay.name

Stochastic parrotology

danmackinlay.name·37w

Deep linear networks

danmackinlay.name·40w

Agent foundations

danmackinlay.name·40w

Coincidences and computation

danmackinlay.name·40w

Discretizing and quantizing neural nets

danmackinlay.name·42w

Bayes predictive framework

danmackinlay.name·44w

Degrees of freedom in NNs

danmackinlay.name·46w

Inductive biases

danmackinlay.name·47w

Headaches

danmackinlay.name·48w

Multi-level agency

danmackinlay.name·48w

Utility and fitness

danmackinlay.name·49w

Utility and fitness

danmackinlay.name·49w·Hacker News

The deep history of intelligence

danmackinlay.name·49w

AI Risk Research Idea - Human Reward Hacking

danmackinlay.name·49w

AI Risk Research Idea - Human Reward Hacking

danmackinlay.name·49w

AI persuasion, AI companionship

danmackinlay.name·49w

Fractal and self-similar behaviour in neural networks

danmackinlay.name·49w

Distributed community NN training

danmackinlay.name·50w

Estimating the Local Learning coefficient

danmackinlay.name·50w

Learning Schrödinger bridges

danmackinlay.name·50w

Log in to enable infinite scrolling