Theoretical computer scientist working on the mathematical foundations of AI alignment.
Toronto, Ontario
Joined July 2015
- Tweets848
- Following800
- Followers131
- Likes3.4K
Vinayak Pathak retweeted
I'm excited to co-found this new AI safety nonprofit research institute in Cambridge, MA, with an amazing team. Join us in making AI safe by design.
resi.org
The Institute for Responsible Superintelligence (RESI): a nonprofit founded by Shafi Goldwasser, Adam Kalai, Vinod Vaikuntanathan. @ShafiGoldwasser @Vinod_MIT @AdamFungi
resi.org
Our goal: build scientific foundations needed to make superintelligence safe by design
Vinayak Pathak retweeted
I don’t think it is rational for anyone to be doing capabilities research at a frontier lab right now. We are not in a Prisoners Dilemma: the situation is very dangerous, and if one person or lab stops it makes it easier and more peer-compatible for other people or labs to stop.
Vinayak Pathak retweeted
Adding “tolerance” to the adversarial PAC learning framework offers new insights on why “perturb and smooth” is a natural method for adversarial learning
Adversarially Robust Learning with Tolerance. (arXiv:2203.00849v1 [stat.ML]) ift.tt/soXVPIR
Vinayak Pathak retweeted
"Black-box Certification and Learning under Adversarial Perturbations" #icml2020
arxiv.org/pdf/2006.16520.pdf
(joint work with Vinayak Pathak and Ruth Urner)
Vinayak Pathak retweeted
Your calendar is like a bank account where you get 168 hours in allowance every week.
Vinayak Pathak retweeted
Beautiful News: The most recent gorgeous graphics on underreported but substantial examples of ongoing progress. informationisbeautiful.net/b…
Vinayak Pathak retweeted
One of my most controversial software opinions is that your sleep quality and stress level matter far, far more than the languages you use or the practices you follow. Nothing else comes close: not type systems, not TDD, not formal methods, not ANYTHING.
Allow me to explain why.
Vinayak Pathak retweeted
An OpenAI employee printed out this AI-written sample and posted it by the recycling bin: blog.openai.com/better-langu…
Vinayak Pathak retweeted
I don’t understand why so many studies keep contradicting each other in this field and yet no one says hold on let’s take a step back and figure out what the hell is going on.
Vinayak Pathak retweeted
People are really out there shoving paper straws into a plastic cups with plastic tops like they’re actually helping the environment.
Vinayak Pathak retweeted
I'm predicting that 2019 will be a relatively normal and boring year. Remember, you heard it here first.
Vinayak Pathak retweeted
A substantially revised edition of "Rationality: From AI to Zombies" (the book version of @ESYudkowsky's original sequences) is now available — in both paper and electronic forms. intelligence.org/2018/12/15/…
Vinayak Pathak retweeted
If the (Morissette) irony of the situation is not obvious, let me spell it out:
Canada is tripping over itself to hire the best in A.I., yet Canadian officials were denying visas to top African A.I. researchers BECAUSE THEY MIGHT NOT LEAVE CANADA! Should we be so lucky!!
Vinayak Pathak retweeted
The problem is that trying to make the algorithm cause people less pain is in a sense a more unnatural problem then getting it to sell more diapers. Maybe you can solve the problem, but it’s harder and weirder.
Vinayak Pathak retweeted
Just cancelled my travel plan to #NeurIPS and asked a colleague to step in for me for a talk because I still have no update about my visa. This is the nth time in my life that I can't do something simply because some policy doesn't like where I come from.
Vinayak Pathak retweeted
Dear @JustinTrudeau @VectorInst @MILAMontreal @CIFAR_News,
How can Canada lead in AI if we cannot even process visas for top AI researchers gives MONTHS of advance warning?
Please RT.