arXiv subjects

Varun Srivastava

Publications and source records attributed to Varun Srivastava.

At least 19 recordsLinked to original sources

arXiv2026

Practical Tomography of Multi-Time Processes

Characterising multi-time quantum processes is essential for analysing temporally correlated noise and for designing effective control and mitigation strategies. A complete operational description through multi-time process tomography requires an informationally complete set of probes, which necessarily includes non-deterministic intermediate operations. On present-day quantum devices, such operations are commonly implemented using mid-circuit measurements and reset, which are technologically limited and can introduce noise and overhead in terms of ancilla requirement. In this work, we study the minimal ancillary dimension required for complete characterisation of multi-time processes. We show that sequential interactions with a single qubit ancilla can generate an informationally complete family of correlated probes for processes of arbitrary length, without requiring mid-circuit measurements or reset. Our result provides a resource-efficient route for complete multi-time process tomography and establishes that one qubit of coherent ancillary memory suffices for full reconstruction of arbitrary multi-time dynamics.

quant-ph↗

arXiv2026

The Impact of Uncertainty Visualization on Trust in Thematic Maps

Thematic maps are widely used to communicate spatial patterns to non-expert audiences. Although uncertainty is inherent in thematic map data, it is rarely visualized, raising questions about how its inclusion affects trust. Prior work offers mixed perspectives: some argue that uncertainty fosters trust through transparency, while others suggest it may reduce trust by introducing confusion. Yet few empirical studies explicitly measure trust in thematic maps. We conducted a between-subjects experiment (N=161) to evaluate how visualizing uncertainty at varying levels (low, medium, high) influences trust. We find that uncertainty visualization generally reduces trust, with greater reductions observed as uncertainty levels increase. However, maps dominated by low uncertainty do not significantly differ in trust from those with no uncertainty. Moreover, while uncertainty visualization tends to make readers question the accuracy of the data, it appears to have a weaker influence on perceptions of the mapmaker's integrity.

cs.HC↗

arXiv2025

Blind-spots of Randomized Benchmarking Under Temporal Correlations

Randomized benchmarking (RB) is a widely adopted protocol for estimating the average gate fidelity in quantum hardware. However, its standard formulation relies on the assumption of temporally uncorrelated noise, an assumption often violated in current devices. In this work, we derive analytic expressions for the average sequence fidelity (ASF) in the presence of temporally correlated (non-Markovian) noise with classical memory, including cases where such correlations originate from interactions with a quantum environment. We show how the ASF can be interpreted to extract meaningful benchmarking parameters under such noise and identify classes of interaction Hamiltonians that render temporal correlations completely invisible to RB. We further provide operational criteria for witnessing temporal correlations due to quantum memory through RB experiments. Importantly, while classical correlations may remain undetectable in the ASF data, they can nonetheless significantly affect worst-case errors quantified by the diamond norm, a metric central to fault tolerant quantum computing. In particular, we demonstrate that temporal correlations may suppress worst-case errors highlighting that temporal correlations may not always have detrimental effects on gate performance.

quant-ph↗

arXiv2025

MapIQ: Evaluating Multimodal Large Language Models for Map Question Answering

Recent advancements in multimodal large language models (MLLMs) have driven researchers to explore how well these models read data visualizations, e.g., bar charts, scatter plots. More recently, attention has shifted to visual question answering with maps (Map-VQA). However, Map-VQA research has primarily focused on choropleth maps, which cover only a limited range of thematic categories and visual analytical tasks. To address these gaps, we introduce MapIQ, a benchmark dataset comprising 14,706 question-answer pairs across three map types: choropleth maps, cartograms, and proportional symbol maps spanning topics from six distinct themes (e.g., housing, crime). We evaluate multiple MLLMs using six visual analytical tasks, comparing their performance against one another and a human baseline. An additional experiment examining the impact of map design changes (e.g., altered color schemes, modified legend designs, and removal of map elements) provides insights into the robustness and sensitivity of MLLMs, their reliance on internal geographic knowledge, and potential avenues for improving Map-VQA performance.

cs.CL↗

arXiv2024

Hamiltonian characterisation of multi-time processes with classical memory

A central problem in open quantum systems is the characterization of non-Markovian processes, where an environment retains the memory of its interaction with the system. A key distinction is whether or not this memory can be simulated classically, as this can lead to efficient modelling and noise mitigation. Powerful tools have been developed recently within the process matrix formalism, a framework that conveniently characterizes all multi-time correlations through a sequence of measurements. This leads to a detailed classification of classical and quantum-memory processes and provides operational procedures to distinguish between them. However, these results leave open the question of what type of system-environment interactions lead to classical memory. More generally, process-matrix methods lack a direct connection to joint system-environment evolution, a cornerstone of open-system modelling. In this work, we characterize Hamiltonian and circuit-based models of system-environment interactions leading to classical memory. We show that general time-dependent Hamiltonians with product eigenstates, and where the environment's eigenstates form a time-independent, orthonormal basis, always produce a particular type of classical memory: probabilistic mixtures of unitary processes. Equivalently, these Hamiltonians are characterized as commuting with a complete set of observables on the environment. Additionally, we show that the most general type of classical memory processes can be generated by a quantum circuit in which the system and environment interact through a specific class of controlled unitaries. Our results establish the first strong link between process-matrix methods and traditional Hamiltonian-based approaches to open quantum systems.

quant-ph↗

arXiv2024

Compressing Large Language Models using Low Rank and Low Precision Decomposition

The prohibitive sizes of Large Language Models (LLMs) today make it difficult to deploy them on memory-constrained edge devices. This work introduces $\rm CALDERA$ -- a new post-training LLM compression algorithm that harnesses the inherent low-rank structure of a weight matrix $\mathbf{W}$ by approximating it via a low-rank, low-precision decomposition as $\mathbf{W} \approx \mathbf{Q} + \mathbf{L}\mathbf{R}$. Here, $\mathbf{L}$ and $\mathbf{R}$ are low rank factors, and the entries of $\mathbf{Q}$, $\mathbf{L}$ and $\mathbf{R}$ are quantized. The model is compressed by substituting each layer with its $\mathbf{Q} + \mathbf{L}\mathbf{R}$ decomposition, and the zero-shot performance of the compressed model is evaluated. Additionally, $\mathbf{L}$ and $\mathbf{R}$ are readily amenable to low-rank adaptation, consequently enhancing the zero-shot performance. $\rm CALDERA$ obtains this decomposition by formulating it as an optimization problem $\min_{\mathbf{Q},\mathbf{L},\mathbf{R}}\lVert(\mathbf{Q} + \mathbf{L}\mathbf{R} - \mathbf{W})\mathbf{X}^\top\rVert_{\rm F}^2$, where $\mathbf{X}$ is the calibration data, and $\mathbf{Q}, \mathbf{L}, \mathbf{R}$ are constrained to be representable using low-precision formats. Theoretical upper bounds on the approximation error of $\rm CALDERA$ are established using a rank-constrained regression framework, and the tradeoff between compression ratio and model performance is studied by analyzing the impact of target rank and quantization bit budget. Results illustrate that compressing LlaMa-$2$ $7$B/$13B$/$70$B and LlaMa-$3$ $8$B models using $\rm CALDERA$ outperforms existing post-training LLM compression techniques in the regime of less than $2.5$ bits per parameter. The implementation is available at: https://github.com/pilancilab/caldera.

cs.LG↗

arXiv2023

Matrix Compression via Randomized Low Rank and Low Precision Factorization

Matrices are exceptionally useful in various fields of study as they provide a convenient framework to organize and manipulate data in a structured manner. However, modern matrices can involve billions of elements, making their storage and processing quite demanding in terms of computational resources and memory usage. Although prohibitively large, such matrices are often approximately low rank. We propose an algorithm that exploits this structure to obtain a low rank decomposition of any matrix $\mathbf{A}$ as $\mathbf{A} \approx \mathbf{L}\mathbf{R}$, where $\mathbf{L}$ and $\mathbf{R}$ are the low rank factors. The total number of elements in $\mathbf{L}$ and $\mathbf{R}$ can be significantly less than that in $\mathbf{A}$. Furthermore, the entries of $\mathbf{L}$ and $\mathbf{R}$ are quantized to low precision formats $--$ compressing $\mathbf{A}$ by giving us a low rank and low precision factorization. Our algorithm first computes an approximate basis of the range space of $\mathbf{A}$ by randomly sketching its columns, followed by a quantization of the vectors constituting this basis. It then computes approximate projections of the columns of $\mathbf{A}$ onto this quantized basis. We derive upper bounds on the approximation error of our algorithm, and analyze the impact of target rank and quantization bit-budget. The tradeoff between compression ratio and approximation accuracy allows for flexibility in choosing these parameters based on specific application requirements. We empirically demonstrate the efficacy of our algorithm in image compression, nearest neighbor classification of image and text embeddings, and compressing the layers of LlaMa-$7$b. Our results illustrate that we can achieve compression ratios as aggressive as one bit per matrix coordinate, all while surpassing or maintaining the performance of traditional compression techniques.

cs.LG↗

arXiv2023

Semi-device-independent certification of quantum non-Markovianity using sequential Random Access Codes

The characterization of multi-time correlations in open quantum systems is of fundamental importance. In this work, we investigate multi-time processes using the process matrix formalism and show that the presence of a quantum non-Markovian environment plays a significant role in enhancing the communication capacity in sequential prepare-transform-measure Quantum Random Access Codes (QRAC). The correlated environment enables a quantum advantage to multiple parties, even with projective measurements. In particular, we show that the Markovian and classical non-Markovian processes, i.e. quantum processes with classical feedback from the environment, do not yield sequential quantum advantage. In contrast, it is possible to achieve an advantage in the presence of a quantum non-Markovian environment. Therefore this approach allows a semi-device-independent certification of quantum non-Markovianity. As opposed to entanglement-detection criteria which require the knowledge of the complete process, this method allows to certify the presence of a quantum non-Markovian environment from the observed measurement statistics. Moreover, quantum memory ameliorates the unambiguous certifiable region of unsharp instruments in a semi-device-independent manner.

quant-ph↗

arXiv2022

Science-Driven Tunable Design of Cosmic Explorer Detectors

Ground-based gravitational-wave detectors like Cosmic Explorer can be tuned to improve their sensitivity at high or low frequencies by tuning the response of the signal extraction cavity. Enhanced sensitivity above 2 kHz enables measurements of the post-merger gravitational-wave spectrum from binary neutron star mergers, which depends critically on the unknown equation of state of hot, ultra-dense matter. Improved sensitivity below 500 Hz favors precision tests of extreme gravity with black hole ringdown signals and improves the detection prospects while facilitating an improved measurement of source properties for compact binary inspirals at cosmological distances. At intermediate frequencies, a more sensitive detector can better measure the tidal properties of neutron stars. We present and characterize the performance of tuned Cosmic Explorer configurations that are designed to optimize detections across different astrophysical source populations. These tuning options give Cosmic Explorer the flexibility to target a diverse set of science goals with the same detector infrastructure. We find that a 40 km Cosmic Explorer detector outperforms a 20 km in all key science goals other than access to post-merger physics. This suggests that Cosmic Explorer should include at least one 40 km facility.

gr-qc↗

arXiv2022

Piezo-deformable Mirrors for Active Mode Matching in Advanced LIGO

The detectors of the laser interferometer gravitational-wave observatory (LIGO) are broadly limited by the quantum noise and rely on the injection of squeezed states of light to achieve their full sensitivity. Squeezing improvement is limited by mode mismatch between the elements of the squeezer and the interferometer. In the current LIGO detectors, there is no way to actively mitigate this mode mismatch. This paper presents a new deformable mirror for wavefront control that meets the active mode matching requirements of advanced LIGO. The active element is a piezo-electric transducer, which actuates on the radius of curvature of a 5 mm thick mirror via an axisymmetric flexure. The operating range of the deformable mirror is 120+-8 mD in vacuum, with an additional 200 mD adjustment range accessible out of vacuum. The scattering into higher-order modes is measured to be <0.2% over the nominal beam radius. These piezo-deformable mirrors meet the stringent noise and vacuum requirements of advanced LIGO and will be used for the next observing run (O4) to control the mode-matching between the squeezer and the interferometer.

astro-ph.IM↗

arXiv2021

1-d Ising model using Kronecker sum and Kronecker product

Calculations in Ising model can be cumbersome and non-intuitive. Here we provide a formulation that addresses these issues for 1-D scenario. We represent the microstates of spin interactions as a diagonal matrix. This is done using two operations: Kronecker sum and Kronecker product. The calculations thus become simple matter of manipulating diagonal matrices. We address the following problems in this work: spins in the magnetic field, open-chain 1-D Ising model, closed-chain 1-D Ising model, 1-D Ising model in an external magnetic field. We believe that this representation will help provide students as well as experts with a simple yet powerful technique to carry out calculations in this model.

cond-mat.stat-mech↗

arXiv2021

A Horizon Study for Cosmic Explorer: Science, Observatories, and Community

This Horizon Study describes a next-generation ground-based gravitational-wave observatory: Cosmic Explorer. With ten times the sensitivity of Advanced LIGO, Cosmic Explorer will push gravitational-wave astronomy towards the edge of the observable universe ($z \sim 100$). The goals of this Horizon Study are to describe and evaluate design concepts for Cosmic Explorer; to plan for the United States' leadership in gravitational-wave astronomy; and to envisage the role of Cosmic Explorer in the international effort to build a "Third-Generation" (3G) observatory network that will make discoveries transformative across astronomy, physics, and cosmology.

astro-ph.IM↗

arXiv2021

High Frame-Rate Phase Camera for High-Resolution Wavefront Sensing in Gravitational-Wave Detectors

We present a novel way of wavefront sensing using a commercially available, continuous-wave time-of-flight camera with QVGA-resolution. This CMOS phase camera is capable of sensing externally modulated light sources with frequencies up to 100 MHz. The high-spatial-resolution of the sensor, combined with our integrated control electronics, allows the camera to image power modulation index as low as -62 dBc/second/pixel. The phase camera is applicable to problems where alignment and mode-mismatch sensing is needed and suited for diagnostic and control applications in gravitational-wave detectors. Specifically, we explore the use of the phase camera in sensing the beat signals due to thermal distortions from point-like heat absorbers on the test masses in the Advanced LIGO detectors. The camera is capable of sensing optical path distortions greater than about two nanometers in the Advanced LIGO input mirrors, limited by the phase resolution. In homodyne readout, the performance can reach up to 0.1 nm, limited by the modulation amplitude sensitivity.

physics.ins-det↗

arXiv2019

Adversarial Approximate Inference for Speech to Electroglottograph Conversion

Speech produced by human vocal apparatus conveys substantial non-semantic information including the gender of the speaker, voice quality, affective state, abnormalities in the vocal apparatus etc. Such information is attributed to the properties of the voice source signal, which is usually estimated from the speech signal. However, most of the source estimation techniques depend heavily on the goodness of the model assumptions and are prone to noise. A popular alternative is to indirectly obtain the source information through the Electroglottographic (EGG) signal that measures the electrical admittance around the vocal folds using dedicated hardware. In this paper, we address the problem of estimating the EGG signal directly from the speech signal, devoid of any hardware. Sampling from the intractable conditional distribution of the EGG signal given the speech signal is accomplished through optimization of an evidence lower bound. This is constructed via minimization of the KL-divergence between the true and the approximated posteriors of a latent variable learned using a deep neural auto-encoder that serves an informative prior. We demonstrate the efficacy of the method at generating the EGG signal by conducting several experiments on datasets comprising multiple speakers, voice qualities, noise settings and speech pathologies. The proposed method is evaluated on many benchmark metrics and is found to agree with the gold standard while proving better than the state-of-the-art algorithms on a few tasks such as epoch extraction.

eess.AS↗

arXiv2019

Detection of Glottal Closure Instants from Raw Speech using Convolutional Neural Networks

Glottal Closure Instants (GCIs) correspond to the temporal locations of significant excitation to the vocal tract occurring during the production of voiced speech. GCI detection from speech signals is a well-studied problem given its importance in speech processing. Most of the existing approaches for GCI detection adopt a two-stage approach (i) Transformation of speech signal into a representative signal where GCIs are localized better, (ii) extraction of GCIs using the representative signal obtained in first stage. The former stage is accomplished using signal processing techniques based on the principles of speech production and the latter with heuristic-algorithms such as dynamic-programming and peak-picking. These methods are thus task-specific and rely on the methods used for representative signal extraction. However, in this paper, we formulate the GCI detection problem from a representation learning perspective where appropriate representation is implicitly learned from the raw-speech data samples. Specifically, GCI detection is cast as a supervised multi-task learning problem solved using a deep convolutional neural network jointly optimizing a classification and regression cost. The learning capability is demonstrated with several experiments on standard datasets. The results compare well with the state-of-the-art algorithms while performing better in the case of presence of real-world non-stationary noise.

cs.SD↗

arXiv2019

Detection Prospects of Core-Collapse Supernovae with Supernova-Optimized Third-Generation Gravitational-wave Detectors

We optimize the third-generation gravitational-wave detector to maximize the range to detect core-collapse supernovae. Based on three-dimensional simulations for core-collapse and the corresponding gravitational-wave waveform emitted, the corresponding detection range for these waveforms is limited to within our galaxy even in the era of third-generation detectors. The corresponding event rate is two per century. We find from the waveforms that to detect core-collapse supernovae with an event rate of one per year, the gravitational-wave detectors need a strain sensitivity of 3$\times10^{-27}~$Hz$^{-1/2}$ in a frequency range from 100~Hz to 1500~Hz. We also explore detector configurations technologically beyond the scope of third-generation detectors. We find with these improvements, the event rate for gravitational-wave observations from CCSN is still low, but is improved to one in twenty years.

gr-qc↗

arXiv2019

Mode matching in GANs through latent space learning and inversion

Generative adversarial networks (GANs) have shown remarkable success in generation of unstructured data, such as, natural images. However, discovery and separation of modes in the generated space, essential for several tasks beyond naive data generation, is still a challenge. In this paper, we address the problem of imposing desired modal properties on the generated space using a latent distribution, engineered in accordance with the modal properties of the true data distribution. This is achieved by training a latent space inversion network in tandem with the generative network using a divergence loss. The latent space is made to follow a continuous multimodal distribution generated by reparameterization of a pair of continuous and discrete random variables. In addition, the modal priors of the latent distribution are learned to match with the true data distribution using minimal-supervision with negligible increment in number of learnable parameters. We validate our method on multiple tasks such as mode separation, conditional generation, and attribute discovery on multiple real world image datasets and demonstrate its efficacy over other state-of-the-art methods.

cs.CV↗

arXiv2018

Toward low-latency coincident precessing and coherent aligned-spin gravitational-wave searches of compact binary coalescences with particle swarm optimization

We investigate the use of particle swarm optimization (PSO) algorithm for detection of gravitational-wave signals from compact binary coalescences. We show that the PSO is fast and effective in searching for gravitational wave signals. The PSO-based aligned-spin coincident multi-detector search recovers appreciably more gravitational-wave signals, for a signal-to-noise ratio (SNR) of 10, the PSO based aligned-spin search recovers approximately 26 $\%$ more events as compared to the template bank searches. The PSO-based aligned-spin coincident search uses 48k matched-filtering operations, and provides a better parameter estimation accuracy at the detection stage, as compared to the PyCBC template-bank search in LIGO's second observation run (O2) with 400k template points. We demonstrate an effective PSO-based precessing coincident search with 320k match-filtering operations per detector. We present results of an all-sky aligned-spin coherent search with 576k match-filtering operations per detector, for some examples of two-, three-, and four-detector networks constituting of the LIGO detectors in Hanford and Livingston, Virgo and KAGRA. Techniques for background estimation that are applicable to real data for PSO-based coincident and coherent searches are also presented.

astro-ph.IM↗