🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Here are the new titles we published in April 25 in Math&Stats, with direct links to the product pages: tinyurl.com/2p9prpfs

#RDDers: check it out. rd2d: Causal Inference in Boundary Discontinuity Designs "This article introduces the R package rd2d, which implements and extends the methodological results developed in Cattaneo et al. (2025) for boundary discontinuity designs [like geographic RDDs]."

Continuous probability distributions "converted "into discrete distributions when using finite precision arithmetic in critical ML applications like differential privacy Gaussians distributions converted as lattice Gaussians: MaxEnt distributions franknielsen.github.io/LatticeGaussia…

#rstats Most applications in R run pretty quickly. But these days, many R users are faced with huge datasets and/or long-running applications, making parallel computation attractive. If you've done this, you likely have used the 'parallel' package, either directly or via

Backward Conformal Prediction. arxiv.org/abs/2505.13732

Very interesting paper on causal inference from statistical models; perhaps more actionable than most causal inference papers: tandfonline.com/doi/full/10.10… #Statistics

Daniel Waxman, Fernando Llorente, Petar M. Djuri\'c. [cs.LG]. Bayesian Ensembling: Insights from Online Optimization and Empirical Bayes. arxiv.org/abs/2505.15638

Article #OTD High dimensional nuisance parameters: an example from parametric survival analysis link.springer.com/article/10.100…

This new working paper provides clear and compelling causal evidence that the flu vaccine is really good for kiddos. "The influenza vaccine reduces outpatient and emergency department visits significantly."

🚨 New preprint 🚨 We introduce Generative Distribution Embeddings (GDEs) — a framework for learning representations of distributions, not just datapoints. GDEs enable multiscale modeling and come with elegant statistical theory and some miraculous geometric results! 🧵

The Modern R Stack for Production AI: doi.org/10.59350/z9frb… Python isn't the only game in town anymore: R can interact with local and cloud LLM APIs, inspect and modify your local R environment and files, implement RAG, computer vision, NLP, evals, & much more #Rstats

This is the state-of-the-art algorithm to calculate confidence intervals for the mean of any bounded random variable (for a fixed number of samples)! Do not miss this one if you like to impress your reviewers with small confidence intervals 😉

🚨 New working paper 🚨 In large cities, wages are higher. But so are inequalities. In fact, low-wage workers earn lower real earnings there. Why? What drives spatial wage disparities? Why some workers work at lower real wages in large cities?

This is a somewhat hard to read paper. But the core idea is simple: adapt standard ideas of conformal testing to test for exchangeability in batch mode, with the difference that conformal test martingales are obtained by compounding e-values. alrw.net/articles/38.pdf

Sihan Chen, Joydeep Chowdhury, Marc G. Genton. [stat.ME]. Robust Maximum $L_q$-Likelihood Covariance Estimation for Replicated Spatial Data. (Replacement). arxiv.org/abs/2407.17592

Congye Wang, Matthew A. Fisher, Heishiro Kanagawa, Wilson Chen, Chris. J. Oates. [stat.CO]. Harnessing the Power of Reinforcement Learning for Adaptive MCMC. arxiv.org/abs/2507.00671

Lisa Gaedke-Merzh\"auser, Vincent Maillou, Fernando Rodriguez Avellaneda, Olaf Schenk, Mathieu Luisier, Paula Moraga, Alexandros Nikolaos Ziogas, H{\aa}vard Rue. [stat.CO]. arxiv.org/abs/2507.06938

*Into the land of automatic differentiation* Material is out! A short PhD course for the CS PhD in Sapienza Università di Roma covering basic and advanced topics in autodiff w/ slides, (rough) Notion notes, and two notebooks including a PyTorch-like implementation. 😅 sscardapane.it/teaching/phd-a…