I am a postdoc in CSAIL at MIT advised by Phillip Isola.
My goal is to uncover the computational and statistical laws of natural and artificial intelligence, and thereby design learning systems that are more efficient, more automatic and more useful in practice.
Reach me at [email protected].
Jeremy Bernstein is a postdoc in CSAIL at MIT advised by Phillip Isola.
His goal is to uncover the computational and statistical laws of natural and artificial intelligence, and thereby design learning systems that are more efficient, more automatic and more useful in practice.
Reach him at [email protected].
News
- 🛝 Slides for my talk at Graphcore on norms and number systems.
- 💾 We released the Modula package along with a paper and docs.
- 🛝 Slides for my talk at Cambridge and Tübingen on AGD.
- 📚 We posted our paper on automatic gradient descent.
- 📚 My PhD thesis is on arXiv.
- 🛝 Slides for my thesis defense.
- 🛝 Slides for my talk at DeepMind.
- 🛝 Slides + 🎥 video for my talk at the Symposium on Algorithmic Information Theory and Machine Learning on the correspondence between neural nets and Bayes Point Machines.
- 📚 Our paper "Investigating Generalization by Controlling Normalized Margin" was accepted to ICML!