arXiv subjects

Takumi Saito

Publications and source records attributed to Takumi Saito.

3 recordsLinked to original sources

arXiv2026

Sparse Sensor Placement for Reducing Forecast Errors in Ensemble Kalman Filtering

Designing efficient observation networks for reducing forecast errors is a fundamental challenge in numerical weather prediction. Data-driven sparse sensor placement (SSP) and ensemble-based data assimilation via the Ensemble Kalman Filter (EnKF) have each addressed this challenge independently, yet their mathematical connections have not been systematically formalized. This study presents a unified theoretical framework integrating SSP and EnKF through optimal experimental design, providing new theoretical and algorithmic results. While conventional SSP methods aim to reduce analysis errors, this study extends the SSP to target forecast error reduction by using a tangent linear model approximated by an ensemble forecast. We derive the Fisher information matrices in the ensemble and model spaces for the EnKF, and clarify the mathematical interpretations of A-, D-, and E-optimality in terms of forecast error reduction. A-optimality in the model space minimizes the mean forecast error variance; D-optimality is ill-defined in the model space due to rank deficiency and is therefore formulated in the ensemble space, where it maximizes the Shannon information content of assimilated observations; and E-optimality in the model space minimizes the worst-case forecast error variance. We further propose a fast greedy algorithm for selecting observation locations under A-optimality in the model space, avoiding matrix inversion at each greedy step and substantially reducing computational cost. Numerical experiments using the Lorenz-96 model support these theoretical findings. Among the three optimality criteria, A-optimality in the model space most consistently reduces forecast spread and root-mean-square error, and yields stable incremental improvements consistent with post-assimilation observation impact diagnostics.

physics.ao-ph↗

arXiv2026

Scale-dependent physical constraints on active intracellular fluctuations

Living cells exhibit nonequilibrium dynamics that shape intracellular processes across length scales, from nanoscale molecular assembly to the organization of macroscopic organelles. While dynamics at micrometer scales are known to be constrained by the actin meshwork at low frequencies, the physical principles governing active fluctuations at the nanoscale remain elusive. Here, we present an analytical framework integrating fluorescence correlation spectroscopy with nonequilibrium modeling to delineate the physical scaling of intracellular mechanics. Applying this framework to fibroblasts, we demonstrate that, in contrast to larger components, nanoscale active fluctuations remain prominent at high frequencies and are predominantly driven by local nonmuscle myosin II activity, establishing a distinct functional hierarchy in intracellular mechanics: local active forces promote rapid spatial exploration for nanoscale molecules, whereas macroscopic actin constraints ensure the structural stability required for larger molecular complexes and organelles. To integrate these scale-dependent behaviors within a single physical framework, we formulated a model that captures the transition of active fluctuations across length scales, revealing that the physical properties of the cytoplasm are governed by the balance between active driving forces and passive structural constraints. Furthermore, applying this model to cellular senescence reveals a reduction in nonequilibrium complexity associated with cytoskeletal rigidification. Thus, our findings bridge the dimensional gap between local molecular kinetics and macroscopic constraints, providing a fundamental physical basis for understanding the hierarchical organization of intracellular dynamics.

physics.bio-ph↗

arXiv2024

LLM-POET: Evolving Complex Environments using Large Language Models

Creating systems capable of generating virtually infinite variations of complex and novel behaviour without predetermined goals or limits is a major challenge in the field of AI. This challenge has been addressed through the development of several open-ended algorithms that can continuously generate new and diverse behaviours, such as the POET and Enhanced-POET algorithms for co-evolving environments and agent behaviour. One of the challenges with existing methods however, is that they struggle to continuously generate complex environments. In this work, we propose LLM-POET, a modification of the POET algorithm where the environment is both created and mutated using a Large Language Model (LLM). By fine-tuning a LLM with text representations of Evolution Gym environments and captions that describe the environment, we were able to generate complex and diverse environments using natural language. We found that not only could the LLM produce a diverse range of environments, but compared to the CPPNs used in Enhanced-POET for environment generation, the LLM allowed for a 34% increase in the performance gain of co-evolution. This increased performance suggests that the agents were able to learn a more diverse set of skills by training on more complex environments.

cs.NE↗