BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Abhishek Agarwal

Abhishek Agarwal appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
hep-thInformation Theorymath.ITmath-phmath.MPquant-phRoboticsApplicationsArtificial IntelligenceComputer Visionmath.STStatistics Theory

Catalog footprint

What is connected

20works
12topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

20 published item(s)

preprint2026arXiv

A Langevin sampler for quantum tomography

Quantum tomography involves obtaining a full classical description of a prepared quantum state from experimental results. We propose a Langevin sampler for quantum tomography, that relies on a new formulation of Bayesian quantum tomography exploiting the Burer-Monteiro factorization of Hermitian positive-semidefinite matrices. If the rank of the target density matrix is known, this formulation allows us to define a posterior distribution that is only supported on matrices whose rank is upper-bounded by the rank of the target density matrix. Conversely, if the target rank is unknown, any upper bound on the rank can be used by our algorithm, and the rank of the resulting posterior mean estimator is further reduced by the use of a low-rank promoting prior density. This prior density is a complex extension of the one proposed in (Annales de l'Institut Henri Poincare Probability and Statistics, 56(2):1465-1483, 2020). We derive a PAC-Bayesian bound on our proposed estimator that matches the best bounds available in the literature, and we show numerically that it leads to strong scalability improvements compared to existing techniques when the rank of the density matrix is known to be small.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

QCalEval: Benchmarking Vision-Language Models for Quantum Calibration Plot Understanding

Quantum computing calibration depends on interpreting experimental data, and calibration plots provide the most universal human-readable representation for this task, yet no systematic evaluation exists of how well vision-language models (VLMs) interpret them. We introduce QCalEval, the first VLM benchmark for quantum calibration plots: 243 samples across 87 scenario types from 22 experiment families, spanning superconducting qubits and neutral atoms, evaluated on six question types in both zero-shot and in-context learning settings. The best general-purpose zero-shot model reaches a mean score of 72.3, and many open-weight models degrade under multi-image in-context learning, whereas frontier closed models improve substantially. A supervised fine-tuning ablation at the 9-billion-parameter scale shows that SFT improves zero-shot performance but cannot close the multimodal in-context learning gap. As a reference case study, we release NVIDIA Ising Calibration 1, an open-weight model based on Qwen3.5-35B-A3B that reaches 74.7 zero-shot average score.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

DURableVS: Data-efficient Unsupervised Recalibrating Visual Servoing via online learning in a structured generative model

Visual servoing enables robotic systems to perform accurate closed-loop control, which is required in many applications. However, existing methods either require precise calibration of the robot kinematic model and cameras or use neural architectures that require large amounts of data to train. In this work, we present a method for unsupervised learning of visual servoing that does not require any prior calibration and is extremely data-efficient. Our key insight is that visual servoing does not depend on identifying the veridical kinematic and camera parameters, but instead only on an accurate generative model of image feature observations from the joint positions of the robot. We demonstrate that with our model architecture and learning algorithm, we can consistently learn accurate models from less than 50 training samples (which amounts to less than 1 min of unsupervised data collection), and that such data-efficient learning is not possible with standard neural architectures. Further, we show that by using the generative model in the loop and learning online, we can enable a robotic system to recover from calibration errors and to detect and quickly adapt to possibly unexpected changes in the robot-camera system (e.g. bumped camera, new objects).

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Group Testing with Runlength Constraints for Topological Molecular Storage

Motivated by applications in topological DNA-based data storage, we introduce and study a novel setting of Non-Adaptive Group Testing (NAGT) with runlength constraints on the columns of the test matrix, in the sense that any two 1's must be separated by a run of at least d 0's. We describe and analyze a probabilistic construction of a runlength-constrained scheme in the zero-error and vanishing error settings, and show that the number of tests required by this construction is optimal up to logarithmic factors in the runlength constraint d and the number of defectives k in both cases. Surprisingly, our results show that runlength-constrained NAGT is not more demanding than unconstrained NAGT when d=O(k), and that for almost all choices of d and k it is not more demanding than NAGT with a column Hamming weight constraint only. Towards obtaining runlength-constrained Quantitative NAGT (QNAGT) schemes with good parameters, we also provide lower bounds for this setting and a nearly optimal probabilistic construction of a QNAGT scheme with a column Hamming weight constraint.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Learning a generative model for robot control using visual feedback

We introduce a novel formulation for incorporating visual feedback in controlling robots. We define a generative model from actions to image observations of features on the end-effector. Inference in the model allows us to infer the robot state corresponding to target locations of the features. This, in turn, guides motion of the robot and allows for matching the target locations of the features in significantly fewer steps than state-of-the-art visual servoing methods. The training procedure for our model enables effective learning of the kinematics, feature structure, and camera parameters, simultaneously. This can be done with no prior information about the robot, structure, and cameras that observe it. Learning is done sample-efficiently and shows strong generalization to test data. Since our formulation is modular, we can modify components of our setup, like cameras and objects, and relearn them quickly online. Our method can handle noise in the observed state and noise in the controllers that we interact with. We demonstrate the effectiveness of our method by executing grasping and tight-fit insertions on robots with inaccurate controllers.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Minimum Hardware Requirements for Hybrid Quantum-Classical DMFT

We numerically emulate noisy intermediate-scale quantum (NISQ) devices and determine the minimal hardware requirements for two-site hybrid quantum-classical dynamical mean-field theory (DMFT). We develop a circuit recompilation algorithm which significantly reduces the number of quantum gates of the DMFT algorithm and find that the quantum-classical algorithm converges if the two-qubit gate fidelities are larger than 99%. The converged results agree with the exact solution within 10%, and perfect agreement within noise-induced error margins can be obtained for two-qubit gate fidelities exceeding 99.9%. By comparison, the quantum-classical algorithm without circuit recompilation requires a two-qubit gate fidelity of at least 99.999% to achieve perfect agreement with the exact solution. We thus find quantum-classical DMFT calculations can be run on the next generation of NISQ devices if combined with the recompilation techniques developed in this work.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Bounds on the Rate of Linear Locally Repairable Codes over Small Alphabets

Locally repairable codes (LRC) have recently been a subject of intense research due to theoretical appeal and their application in distributed storage systems. In an LRC, any coordinate of a codeword can be recovered by accessing only few other coordinates. For LRCs over small alphabet (such as binary), the optimal rate-distance trade-off is unknown. In this paper we provide the tightest known upper bound on the rate of linear LRCs of a given relative distance, an improvement over any previous result, in particular \cite{cadambe2013upper}.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Local Partial Clique Covers for Index Coding

Index coding, or broadcasting with side information, is a network coding problem of most fundamental importance. In this problem, given a directed graph, each vertex represents a user with a need of information, and the neighborhood of each vertex represents the side information availability to that user. The aim is to find an encoding to minimum number of bits (optimal rate) that, when broadcasted, will be sufficient to the need of every user. Not only the optimal rate is intractable, but it is also very hard to characterize with some other well-studied graph parameter or with a simpler formulation, such as a linear program. Recently there have been a series of works that address this question and provide explicit schemes for index coding as the optimal value of a linear program with rate given by well-studied properties such as local chromatic number or partial clique-covering number. There has been a recent attempt to combine these existing notions of local chromatic number and partial clique covering into a unified notion denoted as the local partial clique cover (Arbabjolfaei and Kim, 2014). We present a generalized novel upper-bound (encoding scheme) - in the form of the minimum value of a linear program - for optimal index coding. Our bound also combines the notions of local chromatic number and partial clique covering into a new definition of the local partial clique cover, which outperforms both the previous bounds, as well as beats the previous attempt to combination. Further, we look at the upper bound derived recently by Thapa et al., 2015, and extend their $n$-$\mathsf{GIC}$ (Generalized Interlinked Cycle) construction to $(k,n)$-$\mathsf{GIC}$ graphs, which are a generalization of $k$-partial cliques.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Security in Locally Repairable Storage

In this paper we extend the notion of {\em locally repairable} codes to {\em secret sharing} schemes. The main problem that we consider is to find optimal ways to distribute shares of a secret among a set of storage-nodes (participants) such that the content of each node (share) can be recovered by using contents of only few other nodes, and at the same time the secret can be reconstructed by only some allowable subsets of nodes. As a special case, an eavesdropper observing some set of specific nodes (such as less than certain number of nodes) does not get any information. In other words, we propose to study a locally repairable distributed storage system that is secure against a {\em passive eavesdropper} that can observe some subsets of nodes. We provide a number of results related to such systems including upper-bounds and achievability results on the number of bits that can be securely stored with these constraints.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Fermions, Mass-Gap and Landau Levels: Gauge invariant Hamiltonian for QCD in D=2+1

A gauge-invariant reformulation of QCD in three spacetime dimensions is presented within a Hamiltonian formalism, extending previous work to include fermion fields in the adjoint and fundamental representations. A priori there are several ways to define the gauge-invariant versions of the fermions; a consistent prescription for choosing the fermionic variables is presented. The fermionic contribution to the volume element of the gauge orbit space and the gluonic mass-gap is computed exactly and this contribution is shown to be closely related to the mechanism for induction of Chern-Simons terms by parity-odd fermions. The consistency of the Hamiltonian scheme with known results on index theorems, Landau Levels and renormalization of Chern-Simons level numbers is shown in detail. We also comment on the fermionic contribution to the volume element in relation to issues of confinement and screening.

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

Scattering Amplitudes of Massive N=2 Gauge Theories in Three Dimensions

We study the scattering amplitudes of mass-deformed Chern-Simons theories and Yang-Mills-Chern-Simons theories with N=2 supersymmetry in three dimensions. In particular, we derive the on-shell supersymmetry algebras which underlie the scattering matrices of these theories. We then compute various 3 and 4-point on-shell tree-level amplitudes in these theories. For the mass-deformed Chern-Simons theory, odd-point amplitudes vanish and we find that all of the 4-point amplitudes can be encoded elegantly in superamplitudes. For the Yang-Mills-Chern-Simons theory, we obtain all of the 4-point tree-level amplitudes using a combination of perturbative techniques and algebraic constraints and we comment on difficulties related to computing amplitudes with external gauge fields using Feynman diagrams. Finally, we propose a BCFW recursion relation for mass-deformed theories in three dimensions and discuss the applicability of this proposal to mass-deformed N=2 theories.

0 citations0 reviewsNo code linkedOpen access
preprint2012arXiv

Deconstructing Supersymmetric S-matrices in D <= 2 + 1

Global supersymmetries of the S-matrices of N = 2, 4, 8 supersymmetric Yang-Mills theories in three spacetime dimensions (without matter hypermultiplets) are shown to be SU(1|1), SU(2|2) and SU(2|2) X SU(2|2) respectively. These symmetries are not manifest in the off-shell Lagrangian formulations of these theories. A direct map between these symmetries and their representations in terms of the Yang-Mills degrees of freedom and the corresponding quantities in Chern-Simons-Matter theories with N >= 4 supersymmetry is also obtained. Dimensional reduction of the on-shell observables of the Yang-Mills theories to two spacetime dimensions is also discussed.

0 citations0 reviewsNo code linkedOpen access
preprint2012arXiv

Supersymmetry and Mass Gap in 2+1 Dimensions: A Gauge Invariant Hamiltonian Analysis

A Hamiltonian formulation of Yang-Mills-Chern-Simons theories with $0\leq N\leq 4$ supersymmetry in terms of gauge-invariant variables is presented, generalizing earlier work on nonsupersymmetric gauge theories. Special attention is paid to the volume measure of integration (over the gauge orbit space of the fields) which occurs in the inner product for the wave functions and arguments relating it to the renormalization of the Chern-Simons level number and to mass-gaps in the spectrum of the Hamiltonians are presented. The expression for the integration measure is consistent with the absence of mass gap for theories with extended supersymmetry (in the absence of additional matter hypermultiplets and/or Chern-Simons couplings), while for the minimally supersymmetric case, there is a mass-gap, the scale of which is set by a renormalized level number, in agreement with indications from existing literature. The realization of the supersymmetry algebra and the Hamiltonian in terms of the gauge invariant variables is also presented.

0 citations0 reviewsNo code linkedOpen access
preprint2011arXiv

Hidden Local and Non-local Symmetries of S-matrices of N=2,4,8 SYM in D=2+1

This talk, based principally on arXiv:1103.0786, is devoted to properties of tree-level S-matrices of N=2,4,8 SYM in D=2+1. We'll discuss an on-shell formalism for three-dimensional theories inspired by the spinor-helicity framework in four spacetime dimensions. Our framework will be shown be to particularly well suited for the extraction of hidden symmetries and algebraic structures that the scattering amplitudes of the three-dimensional theories posses. In particular we shall discuss the manifest SO(N) symmetry of the S-matrix to all orders in perturbation theory; a symmetry that the Lagrangians of these theories do not have. After a brief discussion of the ramification of the SO(N) invariance to the D2-M2 brane dualities, we shall introduce an on-shell superfield framework for three-dimensional theories and end with a surprising hint of the existence of non-local symmetries for the S-matrix of the N=8 theory.

0 citations0 reviewsNo code linkedOpen access
preprint2011arXiv

Manifest SO(N) invariance and S-matrices of three-dimensional N=2,4,8 SYM

An on-shell formalism for the computation of S-matrices of SYM theories in three spacetime dimensions is presented. The framework is a generalization of the spinor-helicity formalism in four dimensions. The formalism is applied to establish the manifest SO(N) covariance of the on-shell superalgebra relevant to N =2,4 and 8 SYM theories in d=3. The results are then used to argue for the SO(N) invariance of the S-matrices of these theories: a claim which is proved explicitly for the four-particle scattering amplitudes. Recursion relations relating tree amplitudes of three-dimensional SYM theories are shown to follow from their four-dimensional counterparts. The results for the four-particle amplitudes are verified by tree-level perturbative computations and a unitarity based construction of the integrand corresponding to the leading perturbative correction is also presented for the N=8 theory. For N=8 SYM, the manifest SO(8) symmetry is used to develop a map between the color-ordered amplitudes of the SYM and superconformal Chern-Simons theories, providing a direct connection between on-shell observables of D2 and M2-brane theories.

0 citations0 reviewsNo code linkedOpen access
preprint2011arXiv

Non-local symmetries for Yang-Mills theories and their massive counterparts in two and three dimensions

We identify a non-local symmetry for Yang-Mills theories in 1+1 and 2+1 spacetime dimensions. The symmetry mixes a vector current with the gauge field. The current involved in the symmetry is required to satisfy certain constraints. The explicit solution for the current obeying these constraints, is obtained in two spacetime dimensions and in the abelian case in three dimensions. We conjecture that the current is generated from a non-local gauge and Lorentz invariant mass term in three dimensions and provide some evidence for it. We also posit a conserved current associated with the symmetry generators and derive some of its properties. In the Abelian case, we compute the symmetry algebra and show that additional symmetry generators have to be included for the algebra to close. The algebra contains an SO(2,1) subalgebra. We also comment on the implications of this symmetry for N=1 supersymmetry.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

SU(2|2) for Theories with Sixteen Supercharges at Weak and Strong Coupling

We consider the dimensional reductions of N=4 Supersymmetric Yang-Mills theory on R x S^3 to the three-dimensional theory on R x S^2, the orbifolded theory on R x S^3/Z_k, and the plane-wave matrix model. With explicit emphasis on the three-dimensional theory, we demonstrate the realization of the SU(2|3) algebra in a radial Hamiltonian framework. Using this structure we constrain the form of the spin chains, their S-matrices, and the corresponding one- and two-loop Hamiltonian of the three dimensional theory and find putative signs of integrability up to the two-loop order. The string duals of these theories admit the IIA plane-wave geometry as their Penrose limit. Using known results for strings quantized on this background, we explicitly construct the strong-coupling dual extended SU(2|2) algebra and discuss its implications for the gauge theories.

0 citations0 reviewsNo code linkedOpen access
preprint2009arXiv

A Supersymmetry Preserving Mass-Deformation of N=1 Super Yang-Mills in D=2+1

We construct a massive non-abelian N= 1 SYM theory on R^3. This is achieved by using a non-local gauge and Poincare invariant mass term for gluons due to Nair. The underlying supersymmetry algebra is shown to be a non-central extension of the Poincare algebra by the spacetime rotation group so(3). The incorporation of Chern-Simons couplings in the formalism is also presented. The dimensional reduction of the gauge theory and the SUSY algebra is related to a massive N=2 massive matrix quantum mechanics based on euclidean $Clifford_2(R)$.

0 citations0 reviewsNo code linkedOpen access
preprint2008arXiv

The Hamiltonian Analysis for Yang-Mills Theory on $R\times S^2$

Pure Yang-Mills theory on ${\mathbb R} \times S^2$ is analyzed in a gauge-invariant Hamiltonian formalism. Using a suitable coordinatization for the sphere and a gauge-invariant matrix parametrization for the gauge potentials, we develop the Hamiltonian formalism in a manner that closely parallels previous analysis on ${\mathbb R}^3$. The volume measure on the physical configuration space of the gauge theory, the nonperturbative mass-gap and the leading term of the vacuum wave functional are discussed using a point-splitting regularization. All the results carry over smoothly to known results on ${\mathbb R}^3$ in the limit in which the sphere is de-compactified to a plane.

0 citations0 reviewsNo code linkedOpen access
preprint2004arXiv

The Dilatation Operator of $\cal N$ $=$ 4 SYM and Classical Limits of Spin Chains and Matrix Models

A study of the one loop dilatation operator in the scalar sector of $\cal N$ $=$ 4 SYM is presented. The dilatation operator is analyzed from the point of view of Hamiltonian matrix models. A Lie algebra underlying operator mixing in the planar large $N$ limit is presented, and its role in understanding the integrability of the planar dilatation operator is emphasized. A classical limit of the dilatation operator is obtained by considering a contraction of this Lie algebra, leading to a new way of constructing classical limits for quantum spin chains. An infinite tower of local conserved charges is constructed in this classical limit purely within the context of the matrix model. The deformation of these charges and their relation to the charges of the spin chain is also elaborated upon.

0 citations0 reviewsNo code linkedOpen access