Pointer Theory
Entanglement is just shared mutable state and wavefunction collapse is what happens when we edit that shared state to remove the thing that was measured.
Natural Intelligence
A blog by Sam Greydanus
Latest posts
Moviescape: Google Maps for Cinema
Maps are the solution to doomscrolling. A feed shows you a ranked list and hides everything else whereas a map shows you the whole landscape.
The Cursive Transformer
We train a small GPT-style Transformer model to generate cursive handwriting. The trick to making this work is a custom tokenizer for pen strokes.
Six Experiments in Action Minimization
Using action minimization, we obtain dynamics for six different physical systems including a double pendulum and a gas with a Lennard-Jones potential.
Finding Paths of Least Action with Gradient Descent
The purpose of this simple post is to bring to attention a view of physics which isn’t often communicated in intro courses: the view of physics as optimization.
Studying Growth with Neural Cellular Automata
We train simulated cells to grow into organisms by communicating with their neighbors. Then we use them to study patterns of growth found in nature.
A Structural Optimization Tutorial
Structural optimization lets us design trusses, bridges, and buildings starting from the physics of elastic materials. Let's code it up, from scratch, in 180 lines.
How Simulating the Universe Could Yield Quantum Mechanics
We look at the logistics of simulating the universe. We find that enforcing conservation laws, isotropy, etc. in parallel could lead to quantum-like effects.
Dissipative Hamiltonian Neural Networks
This class of models can learn Hamiltonians from data even when the total energy of the system is not perfectly conserved.
Piecewise-constant Neural ODEs
We propose a timeseries model that can be integrated adaptively. It jumps over simulation steps that are predictable and spends more time on those that are not.