Searcharxiv⌕ Search

arXiv subjects

Apoorva Sharma

Publications and source records attributed to Apoorva Sharma.

31 records · Page 2Linked to original sources
arXiv2022

Deposition of nanosized amino acid functionalized bismuth oxido clusters on gold surfaces

Bismuth compounds are of growing interest with regard to potential applications in catalysis, medicine and electronics, for which their environmentally benign nature is one of the key factors. The most common starting material is bismuth nitrate, which easily hydrolyses to give a large number of condensation products. The so-called bismuth subnitrates are composed of bismuth oxido clusters of varying composition and nuclearity. One reason that hampers the further development of bismuth oxido-based materials, however is the low solubility of the subnitrates, which makes targeted immobilisation on substrates challenging. We present an approach towards solubilisation of bismuth oxido clusters by introducing an amino carboxylate as functional group and a study of the growth mode of these atom-precise nanoclusters on gold surfaces. For this purpose the bismuth oxido cluster [Bi38O45(NO3)20(dmso)28](NO3)4*4dmso (dmso=dimethyl sulfoxide) was reacted with the sodium salt of tert-butyloxycabonyl(Boc)-protected phenylalanine (Phe) to give the soluble and chiral nanocluster [Bi38O45(Boc-Phe)24(dmso)9]. The hydrodynamic diameter of the cluster was estimated with (1.4-1.6) nm (in CH3CN) and (2.2 nm-2.9) nm (in Ethanol) based on dynamic light scattering (DLS). The full exchange of the nitrates by the amino carboxylates was proven by NMR and FTIR as well as elemental analysis (EA) and XPS. The solubility of the bismuth oxido cluster in a protic as well as an aprotic polar organic solvent and the growth mode of the clusters on Au upon spin-, dip-, and drop-coating on gold surfaces were studied. Successful deposition of bismuth oxido cluster was proven by powder XRD, FTIR, and XPS while the microstructure of the resulting films was investigated as a function of the deposition method and the solvent used by SEM, AFM, and optical microscopy.

cond-mat.mtrl-sci↗
arXiv2022

Safe Active Dynamics Learning and Control: A Sequential Exploration-Exploitation Framework

Safe deployment of autonomous robots in diverse scenarios requires agents that are capable of efficiently adapting to new environments while satisfying constraints. In this work, we propose a practical and theoretically-justified approach to maintaining safety in the presence of dynamics uncertainty. Our approach leverages Bayesian meta-learning with last-layer adaptation. The expressiveness of neural-network features trained offline, paired with efficient last-layer online adaptation, enables the derivation of tight confidence sets which contract around the true dynamics as the model adapts online. We exploit these confidence sets to plan trajectories that guarantee the safety of the system. Our approach handles problems with high dynamics uncertainty, where reaching the goal safely is potentially initially infeasible, by first \textit{exploring} to gather data and reduce uncertainty, before autonomously \textit{exploiting} the acquired information to safely perform the task. Under reasonable assumptions, we prove that our framework guarantees the high-probability satisfaction of all constraints at all times jointly, i.e. over the total task duration. This theoretical analysis also motivates two regularizers of last-layer meta-learning models that improve online adaptation capabilities as well as performance by reducing the size of the confidence sets. We extensively demonstrate our approach in simulation and on hardware.

cs.RO↗
arXiv2021

Building on Huang et al. GlossBERT for Word Sense Disambiguation

We propose to take on the problem ofWord Sense Disambiguation (WSD). In language, words of the same form can take different meanings depending on context. While humans easily infer the meaning or gloss of such words by their context, machines stumble on this task.As such, we intend to replicated and expand upon the results of Huang et al.GlossBERT, a model which they design to disambiguate these words (Huang et al.,2019). Specifically, we propose the following augmentations: data-set tweaking(alpha hyper-parameter), ensemble methods, and replacement of BERT with BART andALBERT. The following GitHub repository contains all code used in this report, which extends on the code made available by Huang et al.

cs.CL↗
arXiv2021

On the Problem of Reformulating Systems with Uncertain Dynamics as a Stochastic Differential Equation

We identify an issue in recent approaches to learning-based control that reformulate systems with uncertain dynamics using a stochastic differential equation. Specifically, we discuss the approximation that replaces a model with fixed but uncertain parameters (a source of epistemic uncertainty) with a model subject to external disturbances modeled as a Brownian motion (corresponding to aleatoric uncertainty).

eess.SY↗
arXiv2021

Particle MPC for Uncertain and Learning-Based Control

As robotic systems move from highly structured environments to open worlds, incorporating uncertainty from dynamics learning or state estimation into the control pipeline is essential for robust performance. In this paper we present a nonlinear particle model predictive control (PMPC) approach to control under uncertainty, which directly incorporates any particle-based uncertainty representation, such as those common in robotics. Our approach builds on scenario methods for MPC, but in contrast to existing approaches, which either constrain all or only the first timestep to share actions across scenarios, we investigate the impact of a \textit{partial consensus horizon}. Implementing this optimization for nonlinear dynamics by leveraging sequential convex optimization, our approach yields an efficient framework that can be tuned to the particular information gain dynamics of a system to mitigate both over-conservatism and over-optimism. We investigate our approach for two robotic systems across three problem settings: time-varying, partially observed dynamics; sensing uncertainty; and model-based reinforcement learning, and show that our approach improves performance over baselines in all settings.

eess.SY↗
arXiv2021

Terahertz charge and spin transport in metallic ferromagnets: the role of crystalline and magnetic order

We study the charge and spin dependent scattering in a set of CoFeB thin films whose crystalline order is systematically enhanced and controlled by annealing at increasingly higher temperatures. Terahertz conductivity measurements reveal that charge transport closely follows the development of the crystalline phase, with increasing structural order leading to higher conductivity. The terahertz-induced ultrafast demagnetization, driven by spin-flip scattering mediated by the spin-orbit interaction, is measurable in the pristine amorphous sample and much reduced in the sample with highest crystalline order. Surprisingly, the largest demagnetization is observed at intermediate annealing temperatures, where the enhancement in spin-flip probability is not associated with an increased charge scattering. We are able to correlate the demagnetization amplitude with the magnitude of the in-plane magnetic anisotropy, which we characterize independently, suggesting a magnetoresistance-like description of the phenomenon.

cond-mat.mes-hall↗
arXiv2021

Sketching Curvature for Efficient Out-of-Distribution Detection for Deep Neural Networks

In order to safely deploy Deep Neural Networks (DNNs) within the perception pipelines of real-time decision making systems, there is a need for safeguards that can detect out-of-training-distribution (OoD) inputs both efficiently and accurately. Building on recent work leveraging the local curvature of DNNs to reason about epistemic uncertainty, we propose Sketching Curvature of OoD Detection (SCOD), an architecture-agnostic framework for equipping any trained DNN with a task-relevant epistemic uncertainty estimate. Offline, given a trained model and its training data, SCOD employs tools from matrix sketching to tractably compute a low-rank approximation of the Fisher information matrix, which characterizes which directions in the weight space are most influential on the predictions over the training data. Online, we estimate uncertainty by measuring how much perturbations orthogonal to these directions can alter predictions at a new test input. We apply SCOD to pre-trained networks of varying architectures on several tasks, ranging from regression to classification. We demonstrate that SCOD achieves comparable or better OoD detection performance with lower computational burden relative to existing baselines.

cs.LG↗
arXiv2020

Continuous Meta-Learning without Tasks

Meta-learning is a promising strategy for learning to efficiently learn within new tasks, using data gathered from a distribution of tasks. However, the meta-learning literature thus far has focused on the task segmented setting, where at train-time, offline data is assumed to be split according to the underlying task, and at test-time, the algorithms are optimized to learn in a single task. In this work, we enable the application of generic meta-learning algorithms to settings where this task segmentation is unavailable, such as continual online learning with a time-varying task. We present meta-learning via online changepoint analysis (MOCA), an approach which augments a meta-learning algorithm with a differentiable Bayesian changepoint detection scheme. The framework allows both training and testing directly on time series data without segmenting it into discrete tasks. We demonstrate the utility of this approach on a nonlinear meta-regression benchmark as well as two meta-image-classification benchmarks.

cs.LG↗
arXiv2020

Radiative Decay of Dark Exciton Related Emission in a Sandwiched Monolayer WSe2 Revealed by Room Temperature Micro and Nano Photoluminescence

TMDCs have attracted a lot of attention in recent years due to their unique indirect to direct band gap transition from bulk to monolayer thickness. Strong confinement in the out-of-plane direction enhances the Coulomb potential between the charged particles (e-h pairs) and thus increases the exciton binding energy dramatically. The lattice inversion asymmetry in a monolayer creates two non-equivalent (but degenerate in energy) band edges protected by time reversal polarisation via pseudo-spin. However, the presence of strong spin-orbit coupling in the valence band and weak spin-splitting in the conduction band results in the lowest lying exciton in WX2 (X = S, Se) being spin forbidden and optically dark. Because of their long life times, dark excitons (XD) are highly attractive for quantum optics and optoelectronic applications. To date studying XD emission is limited to cryogenic temperature or required very complex experimental configurations to observe them at room temperature (RT). Here, we demonstrate a novel approach of radiative decay of XD related emissions in 1L-WSe2 studied by micro and nano PL at RT. 1L-WSe2 flakes were sandwiched by noble metal (Au or Ag) substrates and PDMS nano-patches providing a strong local out-of-plane dipole moment with respect to the 2D plane. This strong dipole moment not only enhances the XD in WSe2, it also produces bound excitons due to extrinsic charge defects visible at RT. The spatial distributions of these XD related emissions were studied by TEPL with a spatial resolution < 10 nm confirming the confinement of these excitons within the PDMS nano-patches. Finally, by removing the nano-patches from the top of the flakes we are able to recover the bright excitons in the 1L-WSe2. Our approach paves the way for deep understanding and to harness excitonic properties in low dimensional semiconductors, thus offering a platform towards quantum optics.

cond-mat.mes-hall↗
arXiv2019

Network Offloading Policies for Cloud Robotics: a Learning-based Approach

Today's robotic systems are increasingly turning to computationally expensive models such as deep neural networks (DNNs) for tasks like localization, perception, planning, and object detection. However, resource-constrained robots, like low-power drones, often have insufficient on-board compute resources or power reserves to scalably run the most accurate, state-of-the art neural network compute models. Cloud robotics allows mobile robots the benefit of offloading compute to centralized servers if they are uncertain locally or want to run more accurate, compute-intensive models. However, cloud robotics comes with a key, often understated cost: communicating with the cloud over congested wireless networks may result in latency or loss of data. In fact, sending high data-rate video or LIDAR from multiple robots over congested networks can lead to prohibitive delay for real-time applications, which we measure experimentally. In this paper, we formulate a novel Robot Offloading Problem --- how and when should robots offload sensing tasks, especially if they are uncertain, to improve accuracy while minimizing the cost of cloud communication? We formulate offloading as a sequential decision making problem for robots, and propose a solution using deep reinforcement learning. In both simulations and hardware experiments using state-of-the art vision DNNs, our offloading strategy improves vision task performance by between 1.3-2.6x of benchmark offloading strategies, allowing robots the potential to significantly transcend their on-board sensing accuracy but with limited cost of cloud communication.

cs.RO↗
arXiv2019

Robust and Adaptive Planning under Model Uncertainty

Planning under model uncertainty is a fundamental problem across many applications of decision making and learning. In this paper, we propose the Robust Adaptive Monte Carlo Planning (RAMCP) algorithm, which allows computation of risk-sensitive Bayes-adaptive policies that optimally trade off exploration, exploitation, and robustness. RAMCP formulates the risk-sensitive planning problem as a two-player zero-sum game, in which an adversary perturbs the agent's belief over the models. We introduce two versions of the RAMCP algorithm. The first, RAMCP-F, converges to an optimal risk-sensitive policy without having to rebuild the search tree as the underlying belief over models is perturbed. The second version, RAMCP-I, improves computational efficiency at the cost of losing theoretical guarantees, but is shown to yield empirical results comparable to RAMCP-F. RAMCP is demonstrated on an n-pull multi-armed bandit problem, as well as a patient treatment scenario.

cs.AI↗
arXiv2018

Meta-Learning Priors for Efficient Online Bayesian Regression

Gaussian Process (GP) regression has seen widespread use in robotics due to its generality, simplicity of use, and the utility of Bayesian predictions. The predominant implementation of GP regression is a nonparameteric kernel-based approach, as it enables fitting of arbitrary nonlinear functions. However, this approach suffers from two main drawbacks: (1) it is computationally inefficient, as computation scales poorly with the number of samples; and (2) it can be data inefficient, as encoding prior knowledge that can aid the model through the choice of kernel and associated hyperparameters is often challenging and unintuitive. In this work, we propose ALPaCA, an algorithm for efficient Bayesian regression which addresses these issues. ALPaCA uses a dataset of sample functions to learn a domain-specific, finite-dimensional feature encoding, as well as a prior over the associated weights, such that Bayesian linear regression in this feature space yields accurate online predictions of the posterior predictive density. These features are neural networks, which are trained via a meta-learning (or "learning-to-learn") approach. ALPaCA extracts all prior information directly from the dataset, rather than restricting prior information to the choice of kernel hyperparameters. Furthermore, by operating in the weight space, it substantially reduces sample complexity. We investigate the performance of ALPaCA on two simple regression problems, two simulated robotic systems, and on a lane-change driving task performed by humans. We find our approach outperforms kernel-based GP regression, as well as state of the art meta-learning approaches, thereby providing a promising plug-in tool for many regression tasks in robotics where scalability and data-efficiency are important.

cs.RO↗
arXiv2018

BaRC: Backward Reachability Curriculum for Robotic Reinforcement Learning

Model-free Reinforcement Learning (RL) offers an attractive approach to learn control policies for high-dimensional systems, but its relatively poor sample complexity often forces training in simulated environments. Even in simulation, goal-directed tasks whose natural reward function is sparse remain intractable for state-of-the-art model-free algorithms for continuous control. The bottleneck in these tasks is the prohibitive amount of exploration required to obtain a learning signal from the initial state of the system. In this work, we leverage physical priors in the form of an approximate system dynamics model to design a curriculum scheme for a model-free policy optimization algorithm. Our Backward Reachability Curriculum (BaRC) begins policy training from states that require a small number of actions to accomplish the task, and expands the initial state distribution backwards in a dynamically-consistent manner once the policy optimization algorithm demonstrates sufficient performance. BaRC is general, in that it can accelerate training of any model-free RL algorithm on a broad class of goal-directed continuous control MDPs. Its curriculum strategy is physically intuitive, easy-to-tune, and allows incorporating physical priors to accelerate training without hindering the performance, flexibility, and applicability of the model-free RL algorithm. We evaluate our approach on two representative dynamic robotic learning problems and find substantial performance improvement relative to previous curriculum generation techniques and naive exploration strategies.

cs.RO↗