arXiv subjects

Saurav Singh

Publications and source records attributed to Saurav Singh.

4 recordsLinked to original sources

arXiv2026

Nodal-Surface and Flat-Band Driven Large Anomalous Nernst Effect in Epitaxial Ferromagnetic Weyl Metal Fe5Si3

Magnetic topological materials such as Weyl and Dirac magnets exhibit unconventional electronic properties arising from the interplay between magnetic order and band topology, leading to remarkable thermomagnetic and thermoelectric effects. Here, we investigate the ANE in epitaxial thin films of the Weyl ferromagnet candidate Fe5Si3. A pronounced transverse Nernst response exceeding approximately 1.50 microvolt per kelvin is observed at room temperature, together with a giant anomalous Nernst angle of about 0.56, indicating highly efficient conversion between thermal gradients and transverse electric fields. Beyond the anomalous contribution, a sizable topological Nernst signal of approximately 0.43 microvolt per kelvin persists above room temperature, suggesting the possible presence of real-space Berry curvature associated with nontrivial spin textures. First-principles density functional theory calculations combined with symmetry analysis reveal an unconventional electronic structure in which Weyl nodal lines, nodal surfaces, and nearly flat bands coexist near the Fermi level. This rare concurrence of multiple topological band features produces a strongly enhanced and sharply energy-dependent Berry curvature, which governs both the magnitude and temperature evolution of the observed Nernst response. The close quantitative agreement between calculated anomalous Nernst conductivity and experimental results establishes the topological electronic structure as the dominant origin of the observed thermomagnetic transport, highlighting Fe5Si3 as a chemically simple, low-cost binary topological magnet for exploring both real-space and momentum-space Berry-curvature-driven thermoelectric phenomena.

cond-mat.mtrl-sci↗

arXiv2026

Formulating Reinforcement Learning for Human-Robot Collaboration through Off-Policy Evaluation

Reinforcement learning (RL) has the potential to transform real-world decision-making systems by enabling autonomous agents to learn from experience. Deploying RL in real-world settings, especially in the context of human-robot interaction, requires defining state representations and reward functions, which are critical for learning efficiency and policy performance. Traditional RL approaches often rely on domain expertise and trial-and-error, necessitating extensive human involvement as well as direct interaction with the environment, which can be costly and impractical, especially in complex and safety-critical applications. This work proposes a novel RL framework that leverages off-policy evaluation (OPE) for state space and reward function selection, using only logged interaction data. This approach eliminates the need for real-time access to the environment or human-in-the-loop feedback, greatly reducing the dependency on costly real-time interactions. The proposed approach systematically evaluates multiple candidate state representations and reward functions by training offline RL agents and applying OPE to estimate policy performance. The optimal state space and reward function are selected based on their ability to produce high-performing policies under OPE metrics. Our method is validated on two environments: the Lunar Lander environment by OpenAI Gym, which provides a controlled setting for assessing state space and reward function selection, and a NASA-MATB-II human subjects study environment, which evaluates the approach's real-world applicability to human-robot teaming scenarios. This work enhances the feasibility and scalability of offline RL for real-world environments by automating critical RL design decisions through a data-driven OPE-based evaluation, enabling more reliable, effective, and sustainable RL formulation for complex human-robot interaction settings.

cs.LG↗

arXiv2025

Emergent Anomalous and Topological Hall Responses in an Epitaxial Ferromagnetic Weyl Nodal-Line metal Fe5Si3

The interplay between real and reciprocal space topology yields intrinsically linked transport phenomena in magnetic Weyl systems, wherein the broken time-reversal symmetry, strong Dzyaloshinskii-Moriya interaction, and pronounced uniaxial anisotropy stabilize the momentum-space Berry-curvature monopoles (Weyl nodes) and real-space chiral spin textures. We present a combined first-principles and experimental study of epitaxial Fe5Si3 thin films, establishing them as a magnetic Weyl nodal-line material. First-principles Density Functional Theory (DFT) calculations unambiguously reveal that Fe5Si3 hosts a topologically nontrivial electronic structure containing six pairs of Weyl nodes at or near the Fermi level, accompanied by pronounced Berry curvature at high-symmetry points of the Brillouin Zone. High-quality epitaxial films exhibit robust ferromagnetism with a Curie temperature of ~370 K and strong magneto crystalline anisotropy. The magneto transport measurements on epitaxial films reveal the corresponding Berry curvature-driven responses, including a significantly large intrinsic anomalous Hall conductivity of 504 S/cm and a high anomalous Hall angle of 5.5%, which is in good agreement with DFT calculations. A negative and non-saturating longitudinal magnetoresistance is observed, consistent with a chiral-anomaly contribution from Weyl fermions near the Fermi level (EF). Furthermore, a substantial topological Hall resistivity of 1.6 {\mu}{\Omega} cm robust across a wide temperature range, indicating the possibility of robust chiral spin textures in the thin-film geometry. These combined theoretical and experimental results establish Fe5Si3 as a unique, low-cost, centrosymmetric magnetic Weyl nodal-line material, providing a versatile platform for exploring coupled real and reciprocal space topologies in topological spintronic applications.

cond-mat.mtrl-sci↗

arXiv2023

Spatial and Temporal Attention-based emotion estimation on HRI-AVC dataset

Many attempts have been made at estimating discrete emotions (calmness, anxiety, boredom, surprise, anger) and continuous emotional measures commonly used in psychology, namely `valence' (The pleasantness of the emotion being displayed) and `arousal' (The intensity of the emotion being displayed). Existing methods to estimate arousal and valence rely on learning from data sets, where an expert annotator labels every image frame. Access to an expert annotator is not always possible, and the annotation can also be tedious. Hence it is more practical to obtain self-reported arousal and valence values directly from the human in a real-time Human-Robot collaborative setting. Hence this paper provides an emotion data set (HRI-AVC) obtained while conducting a human-robot interaction (HRI) task. The self-reported pair of labels in this data set is associated with a set of image frames. This paper also proposes a spatial and temporal attention-based network to estimate arousal and valence from this set of image frames. The results show that an attention-based network can estimate valence and arousal on the HRI-AVC data set even when Arousal and Valence values are unavailable per frame.

cs.HC↗