Kevin Waugh

Research Scientist at @SonyAI_global

[Kevin Waugh @kdub0]

@MarlosCMachado https://t.co/FcAhPTXCQR is pretty important. Maybe you’ve read it already

[Kevin Waugh @kdub0]

@SoloGen GPL does not ensure corporations cannot profit off your code. The issue is no one sells software these days. They sell access to services they run. At (some) big corps the only license that will scare them away is AGPL.

[Kevin Waugh @kdub0]

@TZahavy @leanprover Congrats Tom, @vivek_veeriah and team! Hope you’re all doing well.

[Kevin Waugh @kdub0]

@SoloGen @c_voelcker @ChongZitaZhang @AMD Honesty, you probably need a lot fewer than 100 SWE. The thing is it is incredibly risky to build a big piece of software and hope that others will adopt it. This is even the case if you know in advance what you’re building is better than the available alternatives.

[Wen Sun @WenSun1]

The really cool part of REBEL is how it generalizes the classic Natural Policy Gradient algorithm (my favorite RL alg) while being as scalable as REINFORCE :) Some details:

[Tom Zahavy @TZahavy]

We are looking for brilliant and creative candidates with strong programming skills to join us at the Discovery team at @GoogleDeepMind 🧙

We build AI agents that discover new knowledge using RL, planning and LLMs. DM me if you have specific questions about working with us 🙏

[Kevin Waugh @kdub0]

@ancadianadragan @GoogleDeepMind Congrats Anca!

[Alex Kearney @kongaloosh]

I’ve been a bit quiet for the past year!

It's because I’ve been building 🤫

I'm excited to share that the https://t.co/qiR1DCDMIi team is coming out of stealth with a sneak peek.

More to come.

Keep your eyes peeled 👀

[Kevin Waugh @kdub0]

@TmlrOrg I do not appreciate getting an un-declinable review request immediately before Christmas

[Sony AI @SonyAI_global]

As our work in Game AI expands, so does our team. Sony AI is now hiring for a Reinforcement Learning Research Intern to join our global Game AI group. Join us in leveraging AI to develop new, enriching gaming experiences. https://t.co/cOujjJooc8

[James @jmac_ai]

RL PhD students: Want to work on hard RL problems with applications to modern video games? Apply for an internship position at Sony AI! #ReinforcementLearning #AI
https://t.co/L3D6mlwOVU

[Kevin Waugh @kdub0]

@gowerrobert Kinda looks like a proximal version of Frank-Wolfe

[Tom Zahavy @TZahavy]

I'm super excited to share AlphaZeroᵈᵇ, a team of diverse #AlphaZero agents that collaborate to solve #Chess puzzles and demonstrate increased creativity. Check out our paper to learn more!
https://t.co/NveOuYmF6y
A quick 🧵(1/n)

[Kevin Waugh @kdub0]

@marcgbellemare A large number of people already rely on untrustworthy information

[Kevin Waugh @kdub0]

@johnpdickerson @ChrKroer But does it scale?

[Kevin Waugh @kdub0]

@KellenDB It was actually pretty nice weather wise yesterday

[Kevin Waugh @kdub0]

@adad8m @SecPerkinsStan Efficient Projection onto the L1-Ball for Learning in High Dimensions. Duchi et al 2008.

[Kevin Waugh @kdub0]

@SecPerkinsStan @adad8m No. I’m referring to finding max ||x - y||^2 s.t. sum x_i = 1, x >= 0. Using duality you can derived an algorithm for finding the exact solution that runs in O(log n)

[Kevin Waugh @kdub0]

@neurocy @josecalderon Yeah, don't discount the value of leadership though. You're right there are a lot of trade offs though. Job satisfaction and pay are both multi-faceted.