arXiv subjects

Maksim A Kazanskii

Publications and source records attributed to Maksim A Kazanskii.

3 recordsLinked to original sources

arXiv2026

Connectivity-Aware Exploration of Robotic Grasp Spaces

Robotic grasping is typically formulated as the problem of identifying successful actions from a space of candidate grasp poses. However, the organization of successful actions within this space has received less attention. We study the multiscale structure of viable robotic grasps in $SE(3)$ and investigate whether this structure can be exploited for more efficient exploration. Using a large-scale grasp dataset, we show that successful grasp sets exhibit heterogeneous and reproducible connectivity structure across objects. We then introduce a connectivity-aware sampling strategy that incrementally explores the currently observed grasp space by prioritizing potential bridges between components, structural frontiers, boundary extensions, and geometric novelty. In controlled reconstruction experiments, the method recovers the connectivity structure of successful grasp sets substantially more efficiently than random sampling and farthest-point sampling. We further evaluate whether connectivity acquired under hidden grasp viability can improve subsequent grasp discovery, and whether structural experience from previously explored objects can be retrieved and transferred to unseen objects. These results suggest that the spatial organization of viable actions provides information relevant to grasp-space exploration beyond the viability of individual candidate actions. More broadly, they motivate structure-aware exploration as a means of exploiting the geometry of viable action spaces in robotic manipulation.

cs.RO↗

arXiv2026

How to Tame Grokking: Representation Geometry as a Control Signal

Grokking is a phenomenon in which neural networks initially memorize training data and only later exhibit strong generalization after prolonged optimization. Despite extensive recent study, the factors influencing the emergence and timing of grokking remain incompletely understood. We investigate the relationship between representation geometry and delayed generalization. We find that dimensionality collapse consistently precedes the onset of grokking in all evaluated settings. Motivated by these observations, we introduce Geometric Dimensionality Regularization (GeomDR), a simple spectral regularizer that modifies the effective dimensionality of hidden representations during training. Across modular addition, modular division, and permutation composition tasks, GeomDR consistently alters grokking dynamics and can substantially accelerate the onset of generalization depending on the intervention schedule and target dimensionality. In several settings, grokking is accelerated by up to 52 times relative to standard AdamW training. Similar qualitative effects are observed in both multilayer perceptrons and transformers. Together, these results suggest that representation geometry can serve as an effective control signal for grokking and provide evidence that geometric interventions offer a practical approach for studying and influencing delayed generalization in neural networks.

cs.LG↗

arXiv2026

Space-Clock Elevator: Multi-Stage Orbital Transport via Rotating Tethers and Elliptical Nodes

Rotating space tethers have long been proposed as momentum-exchange devices capable of transporting payloads between orbital regimes without continuous propellant expenditure, offering a potential alternative to conventional propulsion for transfers from low Earth orbit to higher orbits. In this work, we numerically investigate a system of multiple rotating tethers distributed across different orbital radii and coupled through intermediate transfer platforms (elliptical nodes) moving along Keplerian trajectories. We identify families of dynamically consistent configurations in which neighboring tethers achieve near-phase synchronization, enabling coordinated payload exchange without impulsive maneuvers. Based on these results, we introduce the concept of a Space-Clock Elevator: a modular orbital transport architecture in which payloads are transferred sequentially between synchronized rotating tethers via intermediate elliptical nodes. Numerical experiments demonstrate that such synchronized tether networks can support outward payload transport while maintaining bounded tether tension and dynamically stable orbital motion.

astro-ph.EP↗