Pooja Rao

Sean Feeney, Pooja Rao, Andreas Klappenecker et al.

In this paper, we present SCALAR (Symbolic Conjecture and LLM-Assisted Reasoning), a neurosymbolic framework for automated conjecture generation in quantum circuit analysis built on top of the CUDA-Q open source framework. The system integrates quantum simulation, symbolic conjecture generation, and LLM-based interpretation. We evaluate SCALAR on 82 MaxCut instances from the MQLib benchmark dataset and extend the analysis to 2,000 randomly generated graphs across four topologies: regular, Erdos-Renyi, Barabasi-Albert, and Watts-Strogatz. The framework generates conjectured bounds relating optimal QAOA parameters to graph invariants, including known relationships such as periodicity constraints on the phase separation parameter $γ$. SCALAR also recovers previously reported parameter transfer phenomena across structurally similar instances. Additionally, the system identifies correlations between graph structural features and optimization landscape properties, which we characterize through invariant-based descriptors. Using CUDA-Q tensor network simulator, we scale experiments to instances of up to 77 qubits. We discuss the accuracy, generality, and limitations of the generated conjectures, including sensitivity to graph class and quantum circuit depth.

Haim Horowitz, Pooja Rao, Santosh Kumar Radha

Synthetic data generation has proven to be a promising solution for addressing data availability issues in various domains. Even more challenging is the generation of synthetic time series data, where one has to preserve temporal dynamics, i.e., the generated time series must respect the original relationships between variables across time. Recently proposed techniques such as generative adversarial networks (GANs) and quantum-GANs lack the ability to attend to the time series specific temporal correlations adequately. We propose using the inherent nature of quantum computers to simulate quantum dynamics as a technique to encode such features. We start by assuming that a given time series can be generated by a quantum process, after which we proceed to learn that quantum process using quantum machine learning. We then use the learned model to generate out-of-sample time series and show that it captures unique and complex features of the learned time series. We also study the class of time series that can be modeled using this technique. Finally, we experimentally demonstrate the proposed algorithm on an 11-qubit trapped-ion quantum machine.

Zitha Sasindran, Harsha Yelchuri, Pooja Rao et al.

We describe a comprehensive methodology for developing user-voice personalized automatic speech recognition (ASR) models by effectively training models on mobile phones, allowing user data and models to be stored and used locally. To achieve this, we propose a resource-aware sub-model-based training approach that considers the RAM, and battery capabilities of mobile phones. By considering the evaluation metric and resource constraints of the mobile phones, we are able to perform efficient training and halt the process accordingly. To simulate real users, we use speakers with various accents. The entire on-device training and evaluation framework was then tested on various mobile phones across brands. We show that fine-tuning the models and selecting the right hyperparameter values is a trade-off between the lowest achievable performance metric, on-device training time, and memory consumption. Overall, our methodology offers a comprehensive solution for developing personalized ASR models while leveraging the capabilities of mobile phones, and balancing the need for accuracy with resource constraints.

Masoud Mohseni, Artur Scherer, K. Grace Johnson et al.

In the span of four decades, quantum computation has evolved from an intellectual curiosity to a potentially realizable technology. Today, small-scale demonstrations have become possible for quantum algorithmic primitives on hundreds of physical qubits and proof-of-principle error-correction on a single logical qubit. Nevertheless, despite significant progress and excitement, the path toward a full-stack scalable technology is largely unknown. There are significant outstanding quantum hardware, fabrication, software architecture, and algorithmic challenges that are either unresolved or overlooked. These issues could seriously undermine the arrival of utility-scale quantum computers for the foreseeable future. Here, we provide a comprehensive review of these scaling challenges. We show how the road to scaling could be paved by adopting existing semiconductor technology to build much higher-quality qubits, employing system engineering approaches, and performing distributed quantum computation within heterogeneous high-performance computing infrastructures. These opportunities for research and development could unlock certain promising applications, in particular, efficient quantum simulation/learning of quantum data generated by natural or engineered quantum systems. To estimate the true cost of such promises, we provide a detailed resource and sensitivity analysis for classically hard quantum chemistry calculations on surface-code error-corrected quantum computers given current, target, and desired hardware specifications based on superconducting qubits, accounting for a realistic distribution of errors. Furthermore, we argue that, to tackle industry-scale classical optimization and machine learning problems in a cost-effective manner, heterogeneous quantum-probabilistic computing with custom-designed accelerators should be considered as a complementary path toward scalability.

Chao Lu, Pooja Rao, Muralikrishnan Gopalakrishnan Meena et al.

The Variational Quantum Linear Solver (VQLS), a hybrid quantum-classical algorithm for solving linear systems, faces a practical scalability bottleneck: the Linear Combination of Unitaries (LCU) decomposition requires O(L^2) circuit evaluations per optimizer iteration, where $L$ can grow as 4^n for n-qubit systems for the worst case scenario. We address this computational bottleneck through two complementary strategies. First, we present a distributed VQLS (D-VQLS) framework, built on NVIDIA CUDA-Q, that enables asynchronous, scalable distribution of the O(L^2) cost-function evaluations. Second, a fast Walsh--Hadamard transform (FWHT)-based Pauli decomposition with 1% coefficient thresholding curbs the exponential growth of LCU terms, reducing L from O}(2^n) to O(1) for n > 6 qubits and compressing the per-iteration circuit complexity from O(n * 4^n) to O(n) for sparse, structured matrices. For a 10-qubit tridiagonal Toeplitz system, this yields a 256x reduction, from 23 million to 90,112 circuits per iteration, while preserving over $99.99\%$ solution fidelity. Additionally, to inform feasibility on early fault-tolerant QPUs, the paper provides resource estimates -- gate counts, qubit requirements, and circuit evaluations per iteration -- for VQLS applied to arbitrary matrices. The D-VQLS framework is validated on the NERSC Perlmutter supercomputer using multi-node, multi-GPU ideal state-vector simulations, achieving over 99.99% fidelity against classical solutions on tridiagonal Toeplitz and Hele--Shaw flow benchmarks, with near-ideal strong scaling up to 24 GPUs and 95.3% weak scaling efficiency at 96 GPUs processing 360,448 circuits per iteration for a 10-qubit system. Systematic profiling identifies the optimal resource allocation for distributed quantum circuit workloads, yielding a 2.52x speedup for the configurations studied.

Elzbieta Gruzewska, Pooja Rao, Sebastien Baur et al.

Radar technology presents untapped potential for continuous, contactless, and passive heart rate monitoring via consumer electronics like mobile phones. However the variety of available radar systems and lack of standardization means that a large new paired dataset collection is required for each radar system. This study demonstrates transfer learning between frequency-modulated continuous wave (FMCW) and impulse-radio ultra-wideband (IR-UWB) radar systems, both increasingly integrated into consumer devices. FMCW radar utilizes a continuous chirp, while IR-UWB radar employs short pulses. Our mm-wave FMCW radar operated at 60 GHz with a 5.5 GHz bandwidth (2.7 cm resolution, 3 receiving antennas [Rx]), and our IR-UWB radar at 8 GHz with a 500 MHz bandwidth (30 cm resolution, 2 Rx). Using a novel 2D+1D ResNet architecture we achieved a mean absolute error (MAE) of 0.85 bpm and a mean absolute percentage error (MAPE) of 1.42% for heart rate monitoring with FMCW radar (N=119 participants, an average of 8 hours per participant). This model maintained performance (under 5 MAE/10% MAPE) across various body positions and heart rate ranges, with a 98.9% recall. We then fine-tuned a variant of this model, trained on single-antenna and single-range bin FMCW data, using a small (N=376, avg 6 minutes per participant) IR-UWB dataset. This transfer learning approach yielded a model with MAE 4.1 bpm and MAPE 6.3% (97.5% recall), a 25% MAE reduction over the IR-UWB baseline. This demonstration of transfer learning between radar systems for heart rate monitoring has the potential to accelerate its introduction into existing consumer devices.

Zitha S, Raghavendra Rao Suresh, Pooja Rao et al.

Training the state-of-the-art speech-to-text (STT) models in mobile devices is challenging due to its limited resources relative to a server environment. In addition, these models are trained on generic datasets that are not exhaustive in capturing user-specific characteristics. Recently, on-device personalization techniques have been making strides in mitigating the problem. Although many current works have already explored the effectiveness of on-device personalization, the majority of their findings are limited to simulation settings or a specific smartphone. In this paper, we develop and provide a detailed explanation of our framework to train end-to-end models in mobile phones. To make it simple, we considered a model based on connectionist temporal classification (CTC) loss. We evaluated the framework on various mobile phones from different brands and reported the results. We provide enough evidence that fine-tuning the models and choosing the right hyperparameter values is a trade-off between the lowest WER achievable, training time on-device, and memory consumption. Hence, this is vital for a successful deployment of on-device training onto a resource-limited environment like mobile phones. We use training sets from speakers with different accents and record a 7.6% decrease in average word error rate (WER). We also report the associated computational cost measurements with respect to time, memory usage, and cpu utilization in mobile phones in real-time.

Preetham Putha, Manoj Tadepalli, Bhargava Reddy et al.

Background: Chest X-rays are the most commonly performed, cost-effective diagnostic imaging tests ordered by physicians. A clinically validated AI system that can reliably separate normals from abnormals can be invaluble particularly in low-resource settings. The aim of this study was to develop and validate a deep learning system to detect various abnormalities seen on a chest X-ray. Methods: A deep learning system was trained on 2.3 million chest X-rays and their corresponding radiology reports to identify various abnormalities seen on a Chest X-ray. The system was tested against - 1. A three-radiologist majority on an independent, retrospectively collected set of 2000 X-rays(CQ2000) 2. Radiologist reports on a separate validation set of 100,000 scans(CQ100k). The primary accuracy measure was area under the ROC curve (AUC), estimated separately for each abnormality and for normal versus abnormal scans. Results: On the CQ2000 dataset, the deep learning system demonstrated an AUC of 0.92(CI 0.91-0.94) for detection of abnormal scans, and AUC(CI) of 0.96(0.94-0.98), 0.96(0.94-0.98), 0.95(0.87-1), 0.95(0.92-0.98), 0.93(0.90-0.96), 0.89(0.83-0.94), 0.91(0.87-0.96), 0.94(0.93-0.96), 0.98(0.97-1) for the detection of blunted costophrenic angle, cardiomegaly, cavity, consolidation, fibrosis, hilar enlargement, nodule, opacity and pleural effusion. The AUCs were similar on the larger CQ100k dataset except for detecting normals where the AUC was 0.86(0.85-0.86). Interpretation: Our study demonstrates that a deep learning algorithm trained on a large, well-labelled dataset can accurately detect multiple abnormalities on chest X-rays. As these systems improve in accuracy, applying deep learning to widen the reach of chest X-ray interpretation and improve reporting efficiency will add tremendous value in radiology workflows and public health screenings globally.

Sasank Chilamkurthy, Rohit Ghosh, Swetha Tanamala et al.

Importance: Non-contrast head CT scan is the current standard for initial imaging of patients with head trauma or stroke symptoms. Objective: To develop and validate a set of deep learning algorithms for automated detection of following key findings from non-contrast head CT scans: intracranial hemorrhage (ICH) and its types, intraparenchymal (IPH), intraventricular (IVH), subdural (SDH), extradural (EDH) and subarachnoid (SAH) hemorrhages, calvarial fractures, midline shift and mass effect. Design and Settings: We retrospectively collected a dataset containing 313,318 head CT scans along with their clinical reports from various centers. A part of this dataset (Qure25k dataset) was used to validate and the rest to develop algorithms. Additionally, a dataset (CQ500 dataset) was collected from different centers in two batches B1 & B2 to clinically validate the algorithms. Main Outcomes and Measures: Original clinical radiology report and consensus of three independent radiologists were considered as gold standard for Qure25k and CQ500 datasets respectively. Area under receiver operating characteristics curve (AUC) for each finding was primarily used to evaluate the algorithms. Results: Qure25k dataset contained 21,095 scans (mean age 43.31; 42.87% female) while batches B1 and B2 of CQ500 dataset consisted of 214 (mean age 43.40; 43.92% female) and 277 (mean age 51.70; 30.31% female) scans respectively. On Qure25k dataset, the algorithms achieved AUCs of 0.9194, 0.8977, 0.9559, 0.9161, 0.9288 and 0.9044 for detecting ICH, IPH, IVH, SDH, EDH and SAH respectively. AUCs for the same on CQ500 dataset were 0.9419, 0.9544, 0.9310, 0.9521, 0.9731 and 0.9574 respectively. For detecting calvarial fractures, midline shift and mass effect, AUCs on Qure25k dataset were 0.9244, 0.9276 and 0.8583 respectively, while AUCs on CQ500 dataset were 0.9624, 0.9697 and 0.9216 respectively.