Matthew Johnson

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

Efficiently computes derivatives of NumPy code.

Probabilistic reasoning and statistical analysis in TensorFlow

Dermatology ddx dataset, Jax implementations of Monte Carlo conformal prediction, plausibility regions and statistical annotation aggregation from our recent work on uncertain ground truth (TMLR'23 and ArXiv pre-print).