This paper has won the Outstanding Paper Award at #ICML2022! 🎉. Thanks again to my collaborators @mirco_mutti and professor Marcello Restelli for making this possible!.

Over the last years we understood how to control generative models via RL/control to maximize expected rewards under KL regularization. But for many applications, that’s not enough. Hence the question: can we control them further?. Excited to present our ✨Oral✨ paper “Flow

Generative models are great at mimicking data — but real (scientific) discovery requires going beyond it. Excited to present our paper “Provable Maximum Entropy Manifold Exploration via Diffusion Models” this Wednesday at ICML 2025!. We propose a scalable, theoretically grounded

RT @vabor112: Fully-funded #PhD in #ML at @EdinburghUni (@InfAtEd, @ancAtEd), in 𝐠𝐞𝐨𝐦𝐞𝐭𝐫𝐢𝐜 𝐥𝐞𝐚𝐫𝐧𝐢𝐧𝐠 and 𝐮𝐧𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲 𝐪𝐮𝐚𝐧𝐭𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧. Applic….

RT @arkrause: 📣We have a tenure-track faculty opening in Responsible AI at @CSatETH:.@ETH_en is a vibrant environme….

RT @SchmidhuberAI: I am hiring 3 postdocs at #KAUST to develop an Artificial Scientist for discovering novel chemical materials for carbon….

RT @tverven: The final draft of @BachFrancis's excellent new learning theory book is available: He doesn't cover VC….

RT @Tim_Roughgarden: Periodic reminder that at my website you can find over 1000 pages of lecture notes on algorithms, data science, game t….

RT @gio_ramponi: @arlet_workshop started now! Come to the first floor to join us #ICML2024

RT @benm_oseley: We just released our ETH Zürich AI in the Sciences and Engineering Master’s course on YouTube! 📚. Prof. Siddhartha Mishra….

Work done with amazing @Fede_Arang, @noahliniger, @mirco_mutti, and @arkrause. Paper:

How can we use RL to estimate an unknown state function by exploiting known environment invariances?. In our #ICML2024 paper (session: Wed 24th, 1.30pm), we present an algorithm that performs active exploration more efficiently by leveraging an abstract representation of the MDP.

Work done with amazing @Manish8Prajapat and @arkrause. Paper:

How can we efficiently (approximately) solve an RL problem with non-additive trajectory rewards?. In our #ICML2024 paper (session: Tue 23rd, 1.30pm), we propose a meta-algorithm that solves this problem by reducing it to a specific sequence of standard MDPs.

RT @ETH_AI_Center: Thrilled to share our 8 conference paper contributions to @icmlconf 2024 next week. Congrats to our doctoral fellows, po….

RT @neu_rips: OK, time for some tweets about distances between Markov chains! Actually this is about a preprint we've just posted on arxiv….

RT @LucaAmb: Work on diffusion! 😜.

RT @gio_ramponi: I am not at #ICLR2024 but @mirco_mutti and @desariky will present our work (w/ @dr_amarx) tomorrow at 16:30! We show how t….

I’ll spend the next 4 days at ICLR. Feel free to reach out if you’d like to chat! . These days I’m obsessed with the interface between decision-making (RL/bandits) and diffusion models and its application in scientific discovery🧪.

RT @mutny_ml: Experiment design, Bayesian optimization or Active learning -- all under one umbrella. The advent of self-driving labs is her….

RT @DavideMaran3: When is it possible to achieve regret guarantees for RL in continuous state-action space? .In our #icml2024 paper we (try….