arXiv subjects

Simon Li

Publications and source records attributed to Simon Li.

7 recordsLinked to original sources

arXiv2025

VOCALoco: Viability-Optimized Cost-aware Adaptive Locomotion

Recent advancements in legged robot locomotion have facilitated traversal over increasingly complex terrains. Despite this progress, many existing approaches rely on end-to-end deep reinforcement learning (DRL), which poses limitations in terms of safety and interpretability, especially when generalizing to novel terrains. To overcome these challenges, we introduce VOCALoco, a modular skill-selection framework that dynamically adapts locomotion strategies based on perceptual input. Given a set of pre-trained locomotion policies, VOCALoco evaluates their viability and energy-consumption by predicting both the safety of execution and the anticipated cost of transport over a fixed planning horizon. This joint assessment enables the selection of policies that are both safe and energy-efficient, given the observed local terrain. We evaluate our approach on staircase locomotion tasks, demonstrating its performance in both simulated and real-world scenarios using a quadrupedal robot. Empirical results show that VOCALoco achieves improved robustness and safety during stair ascent and descent compared to a conventional end-to-end DRL policy

cs.RO↗

arXiv2024

bia-binder: A web-native cloud compute service for the bioimage analysis community

We introduce bia-binder (BioImage Archive Binder), an open-source, cloud-architectured, and web-based coding environment tailored to bioimage analysis that is freely accessible to all researchers. The service generates easy-to-use Jupyter Notebook coding environments hosted on EMBL-EBI's Embassy Cloud, which provides significant computational resources. The bia-binder architecture is free, open-source and publicly available for deployment. It features fast and direct access to images in the BioImage Archive, the Image Data Resource, and the BioStudies databases. We believe that this service can play a role in mitigating the current inequalities in access to scientific resources across academia. As bia-binder produces permanent links to compiled coding environments, we foresee the service to become widely-used within the community and enable exploratory research. bia-binder is built and deployed using helmsman and helm and released under the MIT licence. It can be accessed at binder.bioimagearchive.org and runs on any standard web browser.

q-bio.QM↗

arXiv2024

Impacts of Backside Insulation on the Dynamic On-Resistance of Lateral p-GaN HEMTs-on-Si

We examined the effect of backside insulation on the dynamic on-resistance of lateral p-GaN HEMTs. To gain a comprehensive understanding of the dynamic onresistance difference between substrate grounded and substrate floating p-GaN HEMTs, we conducted in-circuit double pulse testing and long-term direct current (DC) bias stress. We have realized that while backside insulation can enhance the breakdown voltage of lateral p-GaN HEMTs, it also comes with a tradeoff in device reliability. Results through Sentaurus TCAD simulation suggest that the use of backside insulation in devices gradually disperses potential to the buffer barrier. As a result, the potential barrier at the buffer edge of the 2DEG channel decreases significantly, leading to considerable electron trappings at buffer traps. This breakdown voltage and reliability tradeoff also applies to HEMT technologies using insulating substrates.

physics.app-ph↗

arXiv2019

Inferring the Optimal Policy using Markov Chain Monte Carlo

This paper investigates methods for estimating the optimal stochastic control policy for a Markov Decision Process with unknown transition dynamics and an unknown reward function. This form of model-free reinforcement learning comprises many real world systems such as playing video games, simulated control tasks, and real robot locomotion. Existing methods for estimating the optimal stochastic control policy rely on high variance estimates of the policy descent. However, these methods are not guaranteed to find the optimal stochastic policy, and the high variance gradient estimates make convergence unstable. In order to resolve these problems, we propose a technique using Markov Chain Monte Carlo to generate samples from the posterior distribution of the parameters conditioned on being optimal. Our method provably converges to the globally optimal stochastic policy, and empirically similar variance compared to the policy gradient.

cs.LG↗

arXiv2016

Differential operators on the superline, Berezinians, and Darboux transformations

We consider differential operators on a supermanifold of dimension $1|1$. We define non-degenerate operators as those with an invertible top coefficient in the expansion in the "superderivative" $D$ (which is the square root of the shift generator, the partial derivative in an even variable, with the help of an odd indeterminate). They are remarkably similar to ordinary differential operators. We show that every non-degenerate operator can be written in terms of `super Wronskians' (which are certain Berezinians). We apply this to Darboux transformations (DTs), proving that every DT of an arbitrary non-degenerate operator is the composition of elementary first order transformations. Hence every DT corresponds to an invariant subspace of the source operator and, upon a choice of basis in this subspace, is expressed by a super-Wronskian formula. We consider also dressing transformations, i.e., the effect of a DT on the coefficients of the non-degenerate operator. We calculate these transformations in examples and make some general statements.

math.DG↗

arXiv2015

Design Considerations of Biaxially Tensile-Strained Germanium-on-Silicon Lasers

Physical models of Ge energy band structure and material loss were implemented in LASTIP(TM), a 2D simulation tool for edge emitting laser diodes. The model calculation is able to match experimental data available. Important design parameters of a Fabry-Perot Ge laser, such as the cavity length, thickness, width, polycrystalline Si cladding layer thickness were studied and optimized. The laser structure optimizations alone were shown to reduce Ith by 22-fold and increase the differential efficiency by 11 times. The simulations also showed that improving the defect limited carrier lifetime is critical for achieving an efficient and low-threshold Ge laser. With the optimized structure design (300 micron for the cavity length, 0.4 micron for the cavity width, 0.3 micron for the cavity thickness, and 0.6 micron for the polycrystalline Si cladding layer thickness) and a defect limited carrier lifetime of 100 ns, a wall-plug efficiency of 14.6% at 1mW output is predicted, where Jth of 2.8 kA/cm2, Ith of 3.3 mA, I_1mA of 9 mA, and differential efficiency of 23.6% can also be achieved. These are tremendous improvements from the available experimental values at 280 kA/cm2, 756 mA, 837 mA and 1.9%, respectively.

physics.optics↗

arXiv2015

Information-Theoretic Privacy for Smart Metering Systems with a Rechargeable Battery

Smart-metering systems report electricity usage of a user to the utility provider on almost real-time basis. This could leak private information about the user to the utility provider. In this work we investigate the use of a rechargeable battery in order to provide privacy to the user. We assume that the user load sequence is a first-order Markov process, the battery satisfies ideal charge conservation, and that privacy is measured using normalized mutual information (leakage rate) between the user load and the battery output. We consider battery charging policies in this setup that satisfy the feasibility constraints. We propose a series reductions on the original problem and ultimately recast it as a Markov Decision Process (MDP) that can be solved using a dynamic program. In the special case of i.i.d. demand, we explicitly characterize the optimal policy and show that the associated leakage rate can be expressed as a single-letter mutual information expression. In this case we show that the optimal charging policy admits an intuitive interpretation of preserving a certain invariance property of the state. Interestingly an alternative proof of optimality can be provided that does not rely on the MDP approach, but is based on purely information theoretic reductions.

cs.IT↗