Ashish Mehta

Just curious.

[Dalai Lama @DalaiLama]

Time’s always moving on. Nothing can stop it. The question is whether we use our time well or not. We can't do anything about the past, but what happens in the future depends on what we do now. We can create a happier future by remembering that in being human we are all the same.

[Wouter van Amsterdam @WvanAmsterdam]

Bernhard Scholkopf (@bschoelkopf) just published a single author paper titled "Causality for Machine Learning" (https://t.co/Y155IZNqqQ); this should probably at the top of the reading list for many people interested in machine learning / AI; @yudapearl @eliasbareinboim

[Lex Fridman @lexfridman]

Here's Richard Feynman, in 1985, describing Douglas Lenat's heuristic-based system and its winning solutions. Reminds me of reinforcement learning efforts of today. Given objectives & constraints, algorithms may lead to unexpected consequences. Full clip: https://t.co/rRzF9iulqk

[Chip Huyen @chipro]

The move towards bigger models + bigger data makes me worried about the career perspective of for ML researchers. At this point, it seems like you *have* to join a big company that has tremendous resources to make any progress.

[Lex Fridman @lexfridman]

@dwilliams999 "I am not sure that I exist, actually. I am all the writers that I have read, all the people that I have met, all the women that I have loved; all the cities I have visited." - Jorge Luis Borges

[Ashish Mehta @_ashish_m]

❤️🌲? We've got you covered x 20 million. https://t.co/4OHQiblICn #TeamTrees

[Prof. Feynman @ProfFeynman]

"Mathematics is the language in which God has written the universe."

-- Galileo Galilei (1564 - 1642)

[Ashish Mehta @_ashish_m]

@nanjiang_cs @neeraj_wagh

[Ashish Mehta @_ashish_m]

@neeraj_wagh if you actually think about it everything is possible with just the laws of Physics. And hence the infamous "AI Effect" https://t.co/WFQsHCjdIq

[Ashish Mehta @_ashish_m]

.@Nestle you have created a #plasticmonster, end your reliance on throwaway plastic packaging and immediately phase out single-use plastics across your supply chains. #breakfreefromplastic

[Nan Jiang @nanjiang_cs]

My talk at MSR is online now! on our findings and open problems in figuring out minimal assumptions that enable theoretical guarantees for RL. talk was meant to offer a minimalist view of RL accessible to learning theoreticians or even TCS audience. https://t.co/Q6Pt7bDgjF (1/2)

[Prof. Feynman @ProfFeynman]

• Question.

• Hypothesize.

• Experiment.

• Fail.

• Learn.

• Repeat.

[Prof. Feynman @ProfFeynman]

"It is true that a mathematician who is not somewhat of a poet, will never be a perfect mathematician."

- Karl Weierstrass (1815 - 1897)

[François Chollet @fchollet]

Training ever bigger convnets and LSTMs on ever bigger datasets gets us closer to Strong AI -- in the same sense that building taller towers gets us closer to the moon.

[Maxime Chevalier @Love2Code]

An Introduction to Deep Reinforcement Learning (PDF book manuscript, Nov 2018) https://t.co/SG52vhjtXJ

[Ashish Mehta @_ashish_m]

@ilyasut I was wondering what the estimate compute cost was to train this bot

[neuro.social.self @neurosocialself]

Same facts are compatible with any number of different theories .. https://t.co/J1tFTgBWeg

[Maximilian Igl @MaxiIgl]

Our newest work on Multitask Soft Option Learning (MSOL) is out. By leveraging RL as Inference, we are able to overcome several challenges specific to this setting and even learn good option termination policies. With @nntsn @shimon8282, Andrew, Sid and Wendelin

[CBMM @MIT_CBMM]

[MIT News] "Teaching machines to reason about what they see: Researchers combine statistical and symbolic artificial intelligence techniques to speed learning and improve transparency." https://t.co/RbAxzsnZpG @CBMM_MIT 's Josh Tenenbaum

[Julius Rüschenpöhler @rueschenpoehler]

The last days, a fascinating discussion has been happening on #econtwitter & #epitwitter abt #causalinference, pot outcomes, & dir acyclical graphs.

Since #AcademicTwitter is great for open discourse, & bad at keeping all in 1 place, I thought I provide this public good..

1/19