Utkarsh Singhal

Robotics @ Tesla Optimus | Previously: PhD @ UC Berkeley

[Chelsea Finn @chelseabfinn]

Pi models are now running in production settings, in collab with @Ultraroboticsco and @weaverobotics.

We see:
- much higher autonomy with pi-0.6 over using pi-0.5
- fewer mistakes & higher throughput from incorporating data in pre-training

Blog post: https://t.co/ltgLlXRnXg

[Haoru Xue ✈️ CVPR @HaoruXue]

It was absolutely disgraceful seeing the venue people just rudely cut the mic. Even when @breadli428 repeatedly beg for just 3 more minutes and he will be done, the person simply repeated robotically: you should have been done. I was clearing rooms one by one. It’s the event organizer’s fault.

Absolutely no respect at all.

[Utkarsh Singhal @utksinghal]

@antoniloq Congrats! :D

[Utkarsh Singhal @utksinghal]

@_onionesque Also the finding in question is explained in the first ~5-ish pages of the report. They didn’t even read the summary!

[Utkarsh Singhal @utksinghal]

@SwayStar123 Matches my observations too!

One thing I've found interesting is that landscapes/scenes converge the fastest. Is it a coincidence that the first large-ish generative models (e.g., iGANs) were for landscapes?

[Utkarsh Singhal @utksinghal]

Happening now 😀 ICML poster session, E. Hall A-B (E-2203)

[Ryan Feng @ryantfeng]

Q: How do we scale robustness/invariance to foundation models like CLIP?
A: Test-time search! 🔍

Our new work FoCal finds canonical views to boost robustness to complex transforms (e.g. viewpoint): https://t.co/f38zLKu2uz

📍 ICML Poster: Tue 11–1:30, E. Hall A-B (E-2203)
🧵 1/5

[Utkarsh Singhal @utksinghal]

Check out our work @ ICML tomorrow!

I can't join in person 🥲, but I'm genuinely excited to share this work.

FoCal hits the core issues I've encountered working on invariance over the last few years: complex transforms, scaling to foundation models & data-driven invariance 😀

[Ryan Feng @ryantfeng]

Q: How do we scale robustness/invariance to foundation models like CLIP?
A: Test-time search! 🔍

Our new work FoCal finds canonical views to boost robustness to complex transforms (e.g. viewpoint): https://t.co/f38zLKu2uz

📍 ICML Poster: Tue 11–1:30, E. Hall A-B (E-2203)
🧵 1/5

[Utkarsh Singhal @utksinghal]

@aaron_defazio @torchcompiled Would love to read the paper! I have seen similar phenomena with noisy gradient estimates.

My guess before seeing the paper: noise accumulates in the weights over many batches until it starts producing 2nd order (hessian) effects.

[Utkarsh Singhal @utksinghal]

@AllanZhou17 my mute list keeps expanding 🥲

[Utkarsh Singhal @utksinghal]

@EliSennesh Is this even a hot take? Since all EBMs are probabilistic models and vice versa? (The only difference being implementation, which i will conveniently ignore :) )

[Chris Offner @chrisoffner3d]

Our gloriously picked cherries.
Their wicked failure cases.

[Utkarsh Singhal @utksinghal]

Hi all, I'll be at NeurIPS from Tuesday to Sunday.

I would love to chat with people about invariance, test-time optimization, and vision.

Please DM me if you'd like to talk (or catch up)

[Utkarsh Singhal @utksinghal]

@JamesAllingham @kayembruno @shreyaspadhy @JaviAC7 @DavidSKrueger @eric_nalisnick @jmhernandez233 Wonderful work! I love the beautifully drawn graphics especially. :)

I've been working on a similar line of research (https://t.co/x0FdxGRBQr). There is so much exciting stuff to be done. It would be great to chat at NeurIPS!

[Utkarsh Singhal @utksinghal]

@LiyuanLucas Would love to chat too!

[Utkarsh Singhal @utksinghal]

@miniapeur I was told there would be cookies

[Utkarsh Singhal @utksinghal]

@wgussml Really cool observations! Might be related to some of @thisismyhat 's recent work :)

Personally I wonder if this is because (1) the loss functional is convex in function space; (2) averaging gradients ≈ averaging learned functions, at least for MNIST.

[Utkarsh Singhal @utksinghal]

@MaxAifer Sounds exicting!