Tootfinder

Replicable Reinforcement Learning with Linear Function Approximation
Eric Eaton, Marcel Hussing, Michael Kearns, Aaron Roth, Sikata Bela Sengupta, Jessica Sorrell
https://arxiv.org/abs/2509.08660

Replicable Reinforcement Learning with Linear Function Approximation
Replication of experimental results has been a challenge faced by many scientific disciplines, including the field of machine learning. Recent work on the theory of machine learning has formalized replicability as the demand that an algorithm produce identical outcomes when executed twice on different samples from the same distribution. Provably replicable algorithms are especially interesting for reinforcement learning (RL), where algorithms are known to be unstable in practice. While replicab…

Open, Reproducible and Trustworthy Robot-Based Experiments with Virtual Labs and Digital-Twin-Based Execution Tracing
Benjamin Alt, Mareike Picklum, Sorin Arion, Franklin Kenghagho Kenfack, Michael Beetz
https://arxiv.org/abs/2508.11406

Open, Reproducible and Trustworthy Robot-Based Experiments with Virtual Labs and Digital-Twin-Based Execution Tracing
We envision a future in which autonomous robots conduct scientific experiments in ways that are not only precise and repeatable, but also open, trustworthy, and transparent. To realize this vision, we present two key contributions: a semantic execution tracing framework that logs sensor data together with semantically annotated robot belief states, ensuring that automated experimentation is transparent and replicable; and the AICOR Virtual Research Building (VRB), a cloud-based platform for sha…

""[…] There is only one correct metric that should be counted when dealing with software, and that is the user's cognitive load. […] If my Windows/Python/Notepad setup is more ubiquitous, understandable, intuitive and replicable than your obscure Arch/Hyprland build with its hundred painstakingly typed-out customizations for every single software in it, then my setup is better and more minimalist than yours. Full stop. […]""

Crosslisted article(s) found for stat.ML. https://arxiv.org/list/stat.ML/new
[1/1]:
- Replicable Clustering
Hossein Esfandiari, Amin Karbasi, Vahab Mirrokni, Grigoris Velegkas, Felix Zhou

Civil Servants as Builders: Enabling Non-IT Staff to Develop Secure Python and R Tools
Prashant Sharma
https://arxiv.org/abs/2508.07203 https://arxiv.org/p…

Civil Servants as Builders: Enabling Non-IT Staff to Develop Secure Python and R Tools
Current digital government literature focuses on professional in-house IT teams, specialized digital service teams, vendor-developed systems, or proprietary low-code/no-code tools. Almost no scholarship addresses a growing middle ground: technically skilled civil servants outside formal IT roles who can write real code but lack a sanctioned, secure path to deploy their work. This paper introduces a limits-aware, open-source and replicable platform that enables such public servants to develop, p…

Data as Commodity: a Game-Theoretic Principle for Information Pricing
Pasquale Casaburi, Giovanni Piccioli, Pierpaolo Vivo
https://arxiv.org/abs/2510.07101 https://

Data as Commodity: a Game-Theoretic Principle for Information Pricing
Data is the central commodity of the digital economy. Unlike physical goods, it is non-rival, replicable at near-zero cost, and traded under heterogeneous licensing rules. These properties defy standard supply--demand theory and call for new pricing principles. We propose a game-theoretic approach in which the value of a data string emerges from strategic competition among N players betting on an underlying stochastic process, each holding partial information about past outcomes. A better-infor…

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[4/7]:
- Replicable Reinforcement Learning with Linear Function Approximation
Eric Eaton, Marcel Hussing, Michael Kearns, Aaron Roth, Sikata Bela Sengupta, Jessica Sorrell