Tootfinder

Properties of Algorithmic Information Distance
Marcus Hutter
https://arxiv.org/abs/2507.21988 https://arxiv.org/pdf/2507.21988

Properties of Algorithmic Information Distance
The domain-independent universal Normalized Information Distance based on Kolmogorov complexity has been (in approximate form) successfully applied to a variety of difficult clustering problems. In this paper we investigate theoretical properties of the un-normalized algorithmic information distance $d_K$. The main question we are asking in this work is what properties this curious distance has, besides being a metric. We show that many (in)finite-dimensional spaces can(not) be isometrically sc…

Divide, Discover, Deploy: Factorized Skill Learning with Symmetry and Style Priors
Rafael Cathomen, Mayank Mittal, Marin Vlastelica, Marco Hutter
https://arxiv.org/abs/2508.19953

Divide, Discover, Deploy: Factorized Skill Learning with Symmetry and Style Priors
Unsupervised Skill Discovery (USD) allows agents to autonomously learn diverse behaviors without task-specific rewards. While recent USD methods have shown promise, their application to real-world robotics remains underexplored. In this paper, we propose a modular USD framework to address the challenges in the safety, interpretability, and deployability of the learned skills. Our approach employs user-defined factorization of the state space to learn disentangled skill representations. It assig…

Limit-Computable Grains of Truth for Arbitrary Computable Extensive-Form (Un)Known Games
Cole Wyeth, Marcus Hutter, Jan Leike, Jessica Taylor
https://arxiv.org/abs/2508.16245 ht…

Limit-Computable Grains of Truth for Arbitrary Computable Extensive-Form (Un)Known Games
A Bayesian player acting in an infinite multi-player game learns to predict the other players' strategies if his prior assigns positive probability to their play (or contains a grain of truth). Kalai and Lehrer's classic grain of truth problem is to find a reasonably large class of strategies that contains the Bayes-optimal policies with respect to this class, allowing mutually-consistent beliefs about strategy choice that obey the rules of Bayesian inference. Only small classes are known to ha…

How Usable is Automated Feature Engineering for Tabular Data?
Bastian Sch\"afer, Lennart Purucker, Maciej Janowski, Frank Hutter
https://arxiv.org/abs/2508.13932 https://…

How Usable is Automated Feature Engineering for Tabular Data?
Tabular data, consisting of rows and columns, is omnipresent across various machine learning applications. Each column represents a feature, and features can be combined or transformed to create new, more informative features. Such feature engineering is essential to achieve peak performance in machine learning. Since manual feature engineering is expensive and time-consuming, a substantial effort has been put into automating it. Yet, existing automated feature engineering (AutoFE) methods have…

Audio-Vision Contrastive Learning for Phonological Class Recognition
Daiqi Liu, Tom\'as Arias-Vergara, Jana Hutter, Andreas Maier, Paula Andrea P\'erez-Toro
https://arxiv.org/abs/2507.17682

Audio-Vision Contrastive Learning for Phonological Class Recognition
Accurate classification of articulatory-phonological features plays a vital role in understanding human speech production and developing robust speech technologies, particularly in clinical contexts where targeted phonemic analysis and therapy can improve disease diagnosis accuracy and personalized rehabilitation. In this work, we propose a multimodal deep learning framework that combines real-time magnetic resonance imaging (rtMRI) and speech signals to classify three key articulatory dimensio…

Building Forest Inventories with Autonomous Legged Robots -- System, Lessons, and Challenges Ahead
Mat\'ias Mattamala, Nived Chebrolu, Jonas Frey, Leonard Frei{\ss}muth, Haedam Oh, Benoit Casseau, Marco Hutter, Maurice Fallon
https://arxiv.org/abs/2506.20315

Building Forest Inventories with Autonomous Legged Robots -- System, Lessons, and Challenges Ahead
Legged robots are increasingly being adopted in industries such as oil, gas, mining, nuclear, and agriculture. However, new challenges exist when moving into natural, less-structured environments, such as forestry applications. This paper presents a prototype system for autonomous, under-canopy forest inventory with legged platforms. Motivated by the robustness and mobility of modern legged robots, we introduce a system architecture which enabled a quadruped platform to autonomously navigate an…

Towards Provenance-Aware Earth Observation Workflows: the openEO Case Study
H. Omidi, L. Sacco, V. Hutter, G. Irsiegler, M. Claus, M. Schobben, A. Jacob, M. Schramm, S. Fiore
https://arxiv.org/abs/2506.08597

Towards Provenance-Aware Earth Observation Workflows: the openEO Case Study
Capturing the history of operations and activities during a computational workflow is significantly important for Earth Observation (EO). The data provenance helps to collect the metadata that records the lineage of data products, providing information about how data are generated, transferred, manipulated, by whom all these operations are performed and through which processes, parameters, and datasets. This paper presents an approach to improve those aspects, by integrating the data provenance…

Risk-Guided Diffusion: Toward Deploying Robot Foundation Models in Space, Where Failure Is Not An Option
Rohan Thakker, Adarsh Patnaik, Vince Kurtz, Jonas Frey, Jonathan Becktor, Sangwoo Moon, Rob Royce, Marcel Kaufmann, Georgios Georgakis, Pascal Roth, Joel Burdick, Marco Hutter, Shehryar Khattak
https://arxiv.org/abs/2506.1760…

Risk-Guided Diffusion: Toward Deploying Robot Foundation Models in Space, Where Failure Is Not An Option
Safe, reliable navigation in extreme, unfamiliar terrain is required for future robotic space exploration missions. Recent generative-AI methods learn semantically aware navigation policies from large, cross-embodiment datasets, but offer limited safety guarantees. Inspired by human cognitive science, we propose a risk-guided diffusion framework that fuses a fast, learned "System-1" with a slow, physics-based "System-2", sharing computation at both training and inference to couple adaptability …

Learning Accurate Whole-body Throwing with High-frequency Residual Policy and Pullback Tube Acceleration
Yuntao Ma, Yang Liu, Kaixian Qu, Marco Hutter
https://arxiv.org/abs/2506.16986

Learning Accurate Whole-body Throwing with High-frequency Residual Policy and Pullback Tube Acceleration
Throwing is a fundamental skill that enables robots to manipulate objects in ways that extend beyond the reach of their arms. We present a control framework that combines learning and model-based control for prehensile whole-body throwing with legged mobile manipulators. Our framework consists of three components: a nominal tracking policy for the end-effector, a high-frequency residual policy to enhance tracking accuracy, and an optimization-based module to improve end-effector acceleration co…

Retrodicting Chaotic Systems: An Algorithmic Information Theory Approach
Kamal Dingle, Boumediene Hamzi, Marcus Hutter, Houman Owhadi
https://arxiv.org/abs/2507.04780

Retrodicting Chaotic Systems: An Algorithmic Information Theory Approach
Making accurate inferences about data is a key task in science and mathematics. Here we study the problem of \emph{retrodiction}, inferring past values of a series, in the context of chaotic dynamical systems. Specifically, we are interested in inferring the starting value $x_0$ in the series $x_0,x_1,x_2,\dots,x_n$ given the value of $x_n$, and the associated function $f$ which determines the series as $f(x_i)=x_{i+1}$. Even in the deterministic case this is a challenging problem, due to mixin…

Exploring Cosmic Dawn with PANORAMIC I: The Bright End of the UVLF at $z\sim9 -17$
Andrea Weibel, Pascal A. Oesch, Christina C. Williams, Christian Kragh Jespersen, Marko Shuntov, Katherine E. Whitaker, Hakim Atek, Rachel Bezanson, Gabriel Brammer, Iryna Chemerynska, Aidan P. Cloonan, Pratika Dayal, Lukas J. Furtak, Anne Hutter, Zhiyuan Ji, Michael V. Maseda, Mengyuan Xiao
…

Exploring Cosmic Dawn with PANORAMIC I: The Bright End of the UVLF at $z\sim9 -17$
In its first two years of operation, the James Webb Space Telescope has enabled the discovery of a surprising number of UV-bright galaxies at $z\sim10-14$. Their number density is still relatively uncertain, due to cosmic variance effects, and the limited survey area with deep imaging. Here, we combine pure parallel imaging from the PANORAMIC survey with data from legacy fields to constrain the bright end (M$_{\rm UV}<-18.5$) of the UV luminosity function (UVLF) over $0.28\,$deg$^2$ of NIRCam i…

Do-PFN: In-Context Learning for Causal Effect Estimation
Jake Robertson, Arik Reuter, Siyuan Guo, Noah Hollmann, Frank Hutter, Bernhard Sch\"olkopf
https://arxiv.org/abs/2506.06039

Do-PFN: In-Context Learning for Causal Effect Estimation
Estimation of causal effects is critical to a range of scientific disciplines. Existing methods for this task either require interventional data, knowledge about the ground truth causal graph, or rely on assumptions such as unconfoundedness, restricting their applicability in real-world settings. In the domain of tabular machine learning, Prior-data fitted networks (PFNs) have achieved state-of-the-art predictive performance, having been pre-trained on synthetic data to solve tabular prediction…

Experience is the Best Teacher: Grounding VLMs for Robotics through Self-Generated Memory
Guowei Lan, Kaixian Qu, Ren\'e Zurbr\"ugg, Changan Chen, Christopher E. Mower, Haitham Bou-Ammar, Marco Hutter
https://arxiv.org/abs/2507.16713

Feature-Based vs. GAN-Based Learning from Demonstrations: When and Why
Chenhao Li, Marco Hutter, Andreas Krause
https://arxiv.org/abs/2507.05906 https://…

Feature-Based vs. GAN-Based Learning from Demonstrations: When and Why
This survey provides a comparative analysis of feature-based and GAN-based approaches to learning from demonstrations, with a focus on the structure of reward functions and their implications for policy learning. Feature-based methods offer dense, interpretable rewards that excel at high-fidelity motion imitation, yet often require sophisticated representations of references and struggle with generalization in unstructured settings. GAN-based methods, in contrast, use implicit, distributional s…

GraspQP: Differentiable Optimization of Force Closure for Diverse and Robust Dexterous Grasping
Ren\'e Zurbr\"ugg, Andrei Cramariuc, Marco Hutter
https://arxiv.org/abs/2508.15002

GraspQP: Differentiable Optimization of Force Closure for Diverse and Robust Dexterous Grasping
Dexterous robotic hands enable versatile interactions due to the flexibility and adaptability of multi-fingered designs, allowing for a wide range of task-specific grasp configurations in diverse environments. However, to fully exploit the capabilities of dexterous hands, access to diverse and high-quality grasp data is essential -- whether for developing grasp prediction models from point clouds, training manipulation policies, or supporting high-level task planning with broader action options…

Scanner-based real-time automated volumetry reporting of the fetus, amniotic fluid, placenta and umbilical cord for fetal MRI at 0.55T
Sara Neves Silva, Alena Uus, Hadi Waheed, Simi Bansal, Kamilah St Clair, Wendy Norman, Jordina Aviles Verdera, Daniel Cromb, Tomas Woodgate, Milou van Poppel, Johannes K Steinweg, Jacqueline Matthew, Kuberan Pushparajah, David Lloyd, Vanessa Kyriakopoulou, Dimitris Siasakos, Anna David, Joseph V Hajnal, Lisa Story, Mary A Rutherford, Jana Hutter

Scanner-based real-time automated volumetry reporting of the fetus, amniotic fluid, placenta and umbilical cord for fetal MRI at 0.55T
Purpose: This work aims to enable real-time automated intra-uterine volumetric reporting and fetal weight estimation for fetal MRI, deployed directly on the scanner. Methods: A multi-region segmentation nnUNet was trained on 146 bSSFP images of 73 fetal subjects (coronal and axial orientations) for the parcellation of the fetal head, fetal body, placenta, amniotic fluid and umbilical cord from whole uterus bSSFP stacks. A reporting tool was then developed to integrate the segmentation outputs i…

carps: A Framework for Comparing N Hyperparameter Optimizers on M Benchmarks
Carolin Benjamins, Helena Graf, Sarah Segel, Difan Deng, Tim Ruhkopf, Leona Hennig, Soham Basu, Neeratyoy Mallik, Edward Bergman, Deyao Chen, Fran\c{c}ois Cl\'ement, Matthias Feurer, Katharina Eggensperger, Frank Hutter, Carola Doerr, Marius Lindauer
https://

carps: A Framework for Comparing N Hyperparameter Optimizers on M Benchmarks
Hyperparameter Optimization (HPO) is crucial to develop well-performing machine learning models. In order to ease prototyping and benchmarking of HPO methods, we propose carps, a benchmark framework for Comprehensive Automated Research Performance Studies allowing to evaluate N optimizers on M benchmark tasks. In this first release of carps, we focus on the four most important types of HPO task types: blackbox, multi-fidelity, multi-objective and multi-fidelity-multi-objective. With 3 336 tasks…

Multi-critic Learning for Whole-body End-effector Twist Tracking
Aravind Elanjimattathil Vijayan, Andrei Cramariuc, Mattia Risiglione, Christian Gehring, Marco Hutter
https://arxiv.org/abs/2507.08656

Multi-critic Learning for Whole-body End-effector Twist Tracking
Learning whole-body control for locomotion and arm motions in a single policy has challenges, as the two tasks have conflicting goals. For instance, efficient locomotion typically favors a horizontal base orientation, while end-effector tracking may benefit from base tilting to extend reachability. Additionally, current Reinforcement Learning (RL) approaches using a pose-based task specification lack the ability to directly control the end-effector velocity, making smoothly executing trajectori…

Attention-Based Map Encoding for Learning Generalized Legged Locomotion
Junzhe He, Chong Zhang, Fabian Jenelten, Ruben Grandia, Moritz B\"Acher, Marco Hutter
https://arxiv.org/abs/2506.09588

Attention-Based Map Encoding for Learning Generalized Legged Locomotion
Dynamic locomotion of legged robots is a critical yet challenging topic in expanding the operational range of mobile robots. It requires precise planning when possible footholds are sparse, robustness against uncertainties and disturbances, and generalizability across diverse terrains. While traditional model-based controllers excel at planning on complex terrains, they struggle with real-world uncertainties. Learning-based controllers offer robustness to such uncertainties but often lack preci…

Large Scale Robotic Material Handling: Learning, Planning, and Control
Filippo A. Spinelli, Yifan Zhai, Fang Nan, Pascal Egli, Julian Nubert, Thilo Bleumer, Lukas Miller, Ferdinand Hofmann, Marco Hutter
https://arxiv.org/abs/2508.09003

Large Scale Robotic Material Handling: Learning, Planning, and Control
Bulk material handling involves the efficient and precise moving of large quantities of materials, a core operation in many industries, including cargo ship unloading, waste sorting, construction, and demolition. These repetitive, labor-intensive, and safety-critical operations are typically performed using large hydraulic material handlers equipped with underactuated grippers. In this work, we present a comprehensive framework for the autonomous execution of large-scale material handling tasks…

Improving Long-Range Navigation with Spatially-Enhanced Recurrent Memory via End-to-End Reinforcement Learning
Fan Yang, Per Frivik, David Hoeller, Chen Wang, Cesar Cadena, Marco Hutter
https://arxiv.org/abs/2506.05997

Improving Long-Range Navigation with Spatially-Enhanced Recurrent Memory via End-to-End Reinforcement Learning
Recent advancements in robot navigation, especially with end-to-end learning approaches like reinforcement learning (RL), have shown remarkable efficiency and effectiveness. Yet, successful navigation still relies on two key capabilities: mapping and planning, whether explicit or implicit. Classical approaches use explicit mapping pipelines to register ego-centric observations into a coherent map frame for the planner. In contrast, end-to-end learning achieves this implicitly, often through rec…