Nima Reyhani

[Kevin Patrick Murphy @sirbayes]

I am pleased to announce another update to my RL tutorial (https://t.co/fnHAmyg1mH). This time I have added code for RLFT for multi-turn LLM agents, using the awesome Tinker library from @thinkymachines, and the simple ReBN training loop from GEM by @zzlccc et al. With ~100 lines of simple python running on your laptop, you can train an agent based on Qwen3-4B-Instruct to play "guess the number" in 20 minutes.

[Frank Nielsen @FrnkNlsn]

🎓Getting started in information geometry in 1 hour!

- 10-page pdf short introduction (AMS Notices):
👉https://t.co/6hRmU1qR7r

- 45 min. video tutorial:
👉 https://t.co/ONvjDsH8Ag

[Frank Harrell @f2harrell]

I've greatly expanded my chapter on Bayesian clinical trial design with examples of Bayesian power and sample size simulations for time-to-event and ordinal outcomes, incorporating uncertainty in effect size to detect ... https://t.co/jDPYp3i8Gc @vandy_biostat

[Bernhard Schölkopf @bschoelkopf]

Another AI paradox: people are excited about LLMs, some even think that AGI is just around the corner. But some students are depressed how they can still get a PhD. Is it becoming pointless?
Some personal notes on this. (1/8)

[Brendan O'Donoghue @bodonoghue85]

New book draft from @BachFrancis

"Learning Theory from First Principles"
https://t.co/Hg5EY0PmE2

Lots of great material, quite up to date on latest theory.

[Yves Raimond @moustaki]

The proceedings of the #recsys2019 REVEAL workshop now available at: https://t.co/8ytTENBrP1

[Sham Kakade @ShamKakade6]

Also, we recently posted this work https://t.co/bOWAF0UmBr on the theory of policy gradients for reinforcement learning! A long time in the works, this paper finally gets a handle on function approximation with policy gradient methods.

[Ben Recht @beenwrekt]

For folks looking for a thorough intro to the mathematical foundations of reinforcement learning: Video lectures for Bertsekas’ course on RL and control are now available here: https://t.co/cJrYJqX1Rn

[Nicolas Bilodeau @nicobilodeau]

#10yearschallenge

[Mireille Schnitzer @mirschnitzer]

Is your estimator of the average treatment effect simply the best, better than all the rest? Now is your chance to prove it once and for all (or at least until next year). The annual Causal Inference Data Challenge is now live! Datasets available at https://t.co/XO5JSpUibF

[Andrew Gelman et al. @StatModeling]

“The Book of Why” by Pearl and Mackenzie https://t.co/hEMzlDa4wU

[Gurutzeta Guillera @GGuillera]

blockCV: an R package for generating spatially or environmentally separated folds for k‐fold cross‐validation of species distribution models - Valavi - - Methods in Ecology and Evolution - Wiley Online Library https://t.co/uvNPwVsp7R

[Nando de Freitas @NandoDF]

Wow! This is cool

[Quanta Magazine @QuantaMagazine]

Constantinos Daskalakis has been awarded the 2018 Rolf Nevanlinna Prize, one of the highest honors in theoretical computer science, for his explorations of machine learning and core questions in economics. #icm2018 #FieldsMedal2018 https://t.co/0M25htE8Ga

[ICM2018 @ICM_2018]

The winners | Researchers from Germany, India, Iran and Italy take home the 2018 Fields Medal
► https://t.co/7NbpPni7LI

[Luc Sels @LucSels]

The professor who helped make Belgium soccer a world power. @LeuvenU @KU_Leuven in Wall Street Journal.
https://t.co/2oAMCF2gVL

[Pierre-Yves Oudeyer @pyoudeyer]

How many random seeds are needed to compare #DeepRL algorithms?

Our new tutorial to address this key issue of #reproducibility in #reinforcementlearning

PDF: https://t.co/7eHOzhtLuC

Code: https://t.co/0CRRM8RYYr

Blog: https://t.co/rYWM5zPYZB

#machinelearning #neuralnetworks

[James McInerney @mcinerneyj]

do explanations help in recommender systems, and if so, how can they work with bandits? recent slides from a talk @SpotifyResearch during the Spotify ML day in Stockholm https://t.co/81qsSM0k2b