arXiv subjects
Publications and source records attributed to Rahul Singh.
Let $G$ be a connected reductive complex algebraic group and let $q\in\mathbb{C}^{\times}$ be not a root of unity. We prove that every $(G,q)$-connection on a formal punctured disc admits a $(G,q)$-oper structure.
Using the language of q-opers, we propose an algebro-geometric description of the duality between Bethe Ansatz equations for quantum XXZ spin chains and many-body trigonometric Ruijsenaars-Schneider/Macdonald systems for classical groups. Upon this duality, the energy level sets of the classical B/C/D-type Hamiltonians are found to be in bijection with the set of solutions of the XXZ Bethe Ansatz equations of type A, albeit with open boundary conditions. This dictionary generalizes previous results in which a GL(N) XXZ spin chain with twisted periodic boundary conditions was dual to an N-body Ruijsenaars-Schneider system. Our construction implements a $\mathbb{Z}/2\mathbb{Z}$ folding both at the level of the (GL(N),q)-oper data as well as at the level of the N-body trigonometric Ruijsenaars-Schneider model. The nonreduced BC-type systems appear in our analysis as well.
Constructing confidence regions for stochastic gradient descent (SGD) ideally requires estimating the asymptotic covariance matrix, a severe computational bottleneck in high dimensions. Traditional cancellation-based batch means methods bypass this estimation but require inverting a sample batch covariance matrix. This introduces strict mathematical degeneracy when the parameter dimension exceeds the number of batches. To address this problem, we utilize equal batch size batch means method and propose a simultaneous, marginal-friendly framework. The proposed marginal statistics has a asymptotic Student's $t$-distribution, and eliminates the matrix inversion step, entirely circumventing high-dimensional degeneracy. To achieve valid simultaneous coverage, we present an algorithm utilizing wild bootstrap samples drawn from a statistic as a function of only the diagonals of the variance-covariance estimator, and to further incorporate the contribution of cross-dependencies, we introduce an efficient Quasi-Monte Carlo procedure utilizing a $t$-copula approximation. Additionally, we integrate a Lugsail variance estimator to aggressively correct finite-sample bias and under-coverage. The proposed methodology delivers interpretable, simultaneous hyper-rectangular confidence regions that are statistically robust, memory-efficient, and strictly scalable for high-dimensional inference. The theoretical results are supported by extensive numerical simulation analysis through various aspects of dimension, number of batches and error structure.
We study optimal scheduling for remote state estimation over a network with two heterogeneous communication channels: a fast but unreliable channel and a slow but reliable channel. To capture temporal correlations in packet losses, we model the unreliable channel as a Gilbert-Elliott (GE) channel. The remote estimation setup consists of a source, a sensor, and a remote estimator. The source evolves as a discrete-time autoregressive (AR) process, and the sensor decides at each time whether to use the fast unreliable channel or the slow reliable channel. We formulate the scheduling problem faced by the sensor as a Markov decision process (MDP) with a continuous state-space and consider minimizing the infinite horizon average cost criterion, where the cost consists of the squared estimation error and the transmission energy consumed. We establish the existence of an optimal stationary policy. We then characterize the structure of an optimal policy, and show that it has a threshold structure with respect to the estimation error. An optimal policy chooses from amongst the two channels based on whether the error exceeds certain thresholds, where the threshold value depends upon the GE channel state. When the system parameters are unknown, we propose an actor-critic (AC) learning algorithm that exploits the threshold structure of an optimal policy. Numerical results demonstrate that the proposed AC algorithm learns the policy structure effectively and achieves performance close to that of the optimal policy computed using the relative value iteration (RVI).
Retrieval-augmented question answering depends on selecting evidence passages that jointly support answer generation. However, many RAG pipelines rely on top-\(k\) ranking, where passages are selected mainly by individual relevance scores, even though multi-hop questions often require complementary evidence satisfying multiple information requirements. Recent LLM-based selectors address this by treating retrieval as set selection, but using an LLM for this intermediate stage can be costly and difficult to scale. In this work, we formulate evidence selection as a Quadratic Unconstrained Binary Optimization (QUBO) problem. Given a question, candidate passages, and decomposed information requirements, our method constructs an energy function that balances relevance, requirement coverage, support strength, redundancy, complementarity, and compactness. Low-energy solutions correspond to compact evidence subsets that cover the needed requirements while avoiding unnecessary or repetitive context. The selected passages are then passed to a downstream language model for answer generation, separating combinatorial evidence selection from semantic answer generation. We evaluate the proposed QUBO selector on HotpotQA and compare it with LLM-based set selectors and non-LLM baselines including BM25, relevance top-\(k\), maximal marginal relevance, hybrid lexical--semantic ranking, greedy coverage, and random selection. The QUBO selector achieves competitive exact-match and token-F1 performance relative to LLM-based selectors while providing a solver-compatible formulation for structured evidence selection. These results suggest that multi-hop evidence selection can be cast as discrete optimization, opening a path toward RAG pipelines where LLMs are reserved for semantic processing and answer generation, while context selection is handled by Ising/QUBO-compatible solvers.
We introduce a graph bispectrum formulation for characterizing higher-order interactions in graph signals. While conventional graph spectral methods capture only second-order structure, many graph signals exhibit nonlinear interactions that are not reflected in covariance or graph power spectra. Motivated by classical higher-order spectral analysis, we define a graph bispectrum tensor based on third-order moments of graph Fourier coefficients and derive a compact graph bicoherence measure that summarizes nonlinear mode interactions in a low-dimensional and scale-invariant form. We establish key properties of the proposed quantities, including vanishing third-order moments for Gaussian graph signals and a dynamical interpretation in terms of nonlinear mode coupling. Experiments on synthetic random graph signals demonstrate that the proposed measures detect complementary nonlinear dependencies even when second-order statistics are similar. We further apply the method to EEG recordings from the CHB-MIT Scalp EEG Database and show that ictal activity exhibits substantially increased nonlinear graph spectral coupling compared to interictal periods. The proposed approach provides an interpretable and computationally efficient tool for higher-order interaction analysis for graph signals.
Time-varying treatment effects, surrogate-identified treatment effects, and mediation effects can all be written as recursive regressions, in which each regression's predicted values become generated outcomes for the next regression. We study how standard causal estimators behave in this setting. Formally, we compare the recursive plug-in, recursive balancing weight, and recursive doubly robust estimators. When every stage is fitted by ordinary least squares (OLS), the three recursive estimators coincide in any finite sample, whether or not the models are correctly specified. As such, estimation by recursively regressing generated outcomes is numerically equivalent to estimation by recursively balancing generated regressors. Under ridge penalisation for the balancing weights, the doubly robust estimator is a backward recursion of stage-wise blends of penalised and OLS regressions. The weight on the recursive OLS regression decays geometrically in the number of time periods. Therefore, the intuition from the cross-sectional setting, where the bias correction moves the estimator towards OLS, applies less and less as the number of time periods increases. For general convex penalties, we derive an identity at each stage.
Semantic mapping is a core service that enables grounded interactions in emerging Extended Reality (XR) applications such as AI assistants. Deploying this capability on mobile XR devices requires a system that is open-vocabulary, real-time, and low-power. Existing approaches are compute-intensive and assume server-class resources. Cloud offloading offers a practical path, but no existing system splits semantic mapping between the device and the cloud, and current approaches do not address how to manage communication, execution, and memory footprint across the device-cloud boundary. We present SemanticXR, the first device-cloud system for real-time, open-vocabulary semantic mapping and querying under XR power, bandwidth, and memory constraints. Our key insight is to elevate semantically identifiable objects to first-class units of system design, governing how the system communicates, executes, and manages memory across the device and the server. Evaluation against a new, aggressive device-cloud baseline shows that object-level system organization improves server-side mapping latency by 2.2x at equivalent semantic quality. Object-level depth-mapping co-design maintains upstream bandwidth under 2.5 Mbps. On the device, an object-level sparse local map with incremental updates and update prioritization enables sub-100 ms query latency for up to 10,000 objects even under network drops, supports tens of thousands of objects within 500 MB memory footprint, and scales downstream bandwidth with map changes rather than total scene size. The system adds only about 2% to idle device power.
In this paper, we derive rates of convergence in the high-dimensional central limit theorem for Polyak--Ruppert averaged iterates generated by entropy-regularized asynchronous Q-learning with linear function approximation and a polynomial stepsize $k^{-\omega}$, $\omega \in (1/2,1)$. Assuming that the sequence of observed triples $(s_k,a_k,s_{k+1})_{k \geq 0}$ forms a uniformly geometrically ergodic Markov chain, and under suitable regularity conditions for the projected soft Bellman equation, we establish a Gaussian approximation bound in the convex distance with rate of order $n^{-1/4}$, up to polylogarithmic factors in $n$, where $n$ is the number of samples used by the algorithm. To obtain this result, we combine a linearization of the soft Bellman recursion with a Gaussian approximation for the leading martingale term. Finally, we derive high-order moment bounds for the algorithm's last iterate, which might be of independent interest.
We study policy gradient methods for reinforcement learning in non-Markovian decision processes (NMDPs), where observations and rewards depend on the entire interaction history. To handle this dependence, the agent maintains an internal state that is recursively updated to provide a compact summary of past observations and actions. In contrast to approaches that treat the agent state dynamics as fixed or learn it via predictive objectives, we propose a reward-centric formulation that jointly optimizes the agent state dynamics and the control policy to maximize the expected cumulative reward. To this end, we consider a class of Agent State-Markov (ASM) policies, comprising an agent state dynamics and a control policy that maps the agent state to actions. We establish a novel policy gradient theorem for ASM policies, extending the classical policy gradient results from the Markovian setting to episodic and infinite-horizon discounted NMDPs. Building on this gradient expression, we propose the Agent State-Markov Policy Gradient (ASMPG) algorithm, which leverages the recursive structure of the agent state dynamics for efficient optimization. We establish finite-time and almost sure convergence guarantees, and empirically demonstrate that, on a range of non-Markovian tasks, ASMPG outperforms baselines that learn state representations via predictive objectives.
Symmetry-protected topological phases of matter, characterized by non-trivial band topology, are spectrally gapped and show non-trivial boundary phenomena. Here, we show that scattering states when interjected by an array of periodically modulated defects can result in emergent topological phases whose properties can be tuned by modulating the defect strengths. We dub this the Su-Schrieffer-Heeger network. We show that a scattering-matrix network model can capture the emergent symmetries and nontrivial winding of the quasienergy bands, which lead to distinct transport signatures and can be further periodically driven to realize a robust Thouless charge pump. We show that a microscopic lattice model embedded with a defect superlattice yields Bloch minibands that directly map to the network problem. We further verify that the physics we report is stable to disorder and point out concrete experimental solid-state platforms where it is readily realizable. Our work, in contrast to engineering atomic Hamiltonians, shows that defect engineering on metallic platforms can lead to emergent topological phases of quantum matter.
Linear stochastic transitivity is a central assumption in paired comparison models that is rarely verified in practice. Empirical violations, however, are common and can substantially affect inference and ranking. We develop a class of tests for detecting lack of fit in cardinal paired comparison models, where lack of fit is characterized by the presence of cyclical preferences among subsets of items. We propose a suite of tests adapted to different regimes governing the growth of the comparison graph. For a fixed number of items, the proposed procedures exhibit substantially improved power relative to the classical Kendall--Smith test and its cardinal analogue. We further extend the framework to high--dimensional, sparse comparison graphs near the connectivity threshold in random graph models. The theoretical analysis characterizes the behavior of the tests under both the null and alternative, with particular emphasis on limits of detectability and consistency. Simulation studies corroborate the theoretical findings, and applications to real data uncover substantial and previously unrecognized intransitivity and structural lack of fit.
We present the first uniform-in-time high-probability bound for SGD under the PL condition, where the gradient noise contains both Markovian and martingale difference components. This significantly broadens the scope of finite-time guarantees, as the PL condition arises in many machine learning and deep learning models while Markovian noise naturally arises in decentralized optimization and online system identification problems. We further allow the magnitude of noise to grow with the function value, enabling the analysis of many practical sampling strategies. In addition to the high-probability guarantee, we establish a matching $1/k$ decay rate for the expected suboptimality. Our proof technique relies on the Poisson equation to handle the Markovian noise and a probabilistic induction argument to address the lack of almost-sure bounds on the objective. Finally, we demonstrate the applicability of our framework by analyzing three practical optimization problems: token-based decentralized linear regression, supervised learning with subsampling for privacy amplification, and online system identification.
Applications in emerging domains such as XR are being built as compound inference systems, where multiple ML models are composed in the form of a task graph to service each request. Serving these compound systems efficiently raises two questions: how to apportion end-to-end latency and accuracy budgets between different tasks in a compound inference system, and how to allocate resources effectively for different models with varying resource requirements. We present JigsawServe, the first serving framework that jointly optimizes for latency, accuracy, and cost in terms of GPU resources by adaptively choosing model variants and performing fine-grained resource allocation by spatially partitioning the GPUs for each task of a compound inference system. Analytical evaluation of a system with a large number of GPUs shows that JigsawServe can increase the maximum serviceable demand (in requests per second) by 11.3x when compared to the closest prior work. Our empirical evaluation shows that for a large range of scenarios, JigsawServe consumes only 43.3% of the available GPU resources while meeting accuracy SLOs with less than 0.6% latency SLO violations. All of the features in JigsawServe contribute to this high efficiency -- sacrificing any one feature of accuracy scaling, GPU spatial partitioning, or task-graph-informed resource budgeting significantly reduces efficiency.
Verification of binary neural network (BNN) robustness is NP-hard, as it can be formulated as a combinatorial search for an adversarial perturbation that induces misclassification. Exact verification methods therefore scale poorly with problem dimension, motivating the use of hardware-accelerated heuristics and unconventional computing platforms, such as Ising solvers, that can efficiently explore complex energy landscapes and discover high-quality solutions. In this work, we reformulate BNN robustness verification as a quadratic unconstrained binary optimization (QUBO) problem and solve it using a digital compute-in-memory (DCIM) SRAM-based Ising machine. Instead of requiring globally optimal solutions, we exploit imperfect solutions produced by the DCIM Ising machine to extract adversarial perturbations and thereby demonstrate the non-robustness of the BNN. The proposed architecture stores quantized QUBO coefficients in approximately 9.1~Mb of SRAM and performs annealing in memory via voltage-controlled pseudo-read dynamics, enabling iterative updates with minimal data movement. Experimental projections indicate that the proposed approach achieves a $178\times$ acceleration in convergence rate and a $1538\times$ improvement in power efficiency relative to conventional CPU-based implementations.
Topological superconductivity could host quasiparticles that are key candidates for fault-tolerant quantum computation due to their immunity to noise as they obey non-Abelian exchange statistics. For example, in the case of Majorana Zero Modes (MZM), braiding enables two topologically protected quantum gates. While their direct manipulation in solid-state systems remains experimentally challenging, digital emulation of MZM behavior has provided insight as well as a deeper understanding of controlling these topological quantum systems. This emulation is typically accomplished by mapping the topological and trivial phases of a Majorana system to ferromagnetic and paramagnetic Hamiltonians of a spin-glass model. This approach usually relies on adiabatic evolution of superconducting Hamiltonians, which require circuits with very large depths. In this work, we present a resource-efficient method to emulate MZM braiding in a trijunction geometry using a quantum processor. We introduce direct braiding operators which simulate the evolution more efficiently, reducing the quantum gate overhead. We then further generalize this method to emulate braiding operations in extended trijunction architectures based on Kitaev chains.
Ranking geographical or administrative units, such as countries or states, is a well-known approach for comparing developmental progress and informing evidence-based policymaking. Existing ranking methodologies typically rely on a single indicator, such as Gross Domestic Product (GDP), or a limited subset of indicators, e.g., the Human Development Index (HDI). However, to the best of our knowledge, a ranking methodology based on a large set of indicator variables is not available in the literature. To address this gap, we present an inclusive ranking methodology. We utilize the Bayesian Bradley-Terry (BT) model, which allows us to incorporate relevant prior information. We model the prior covariance of the BT merit parameters using an independent covariate, such that units with similar covariate values exhibit higher covariance, which decays as differences in the covariate increase. A hybrid of Metropolis-Hastings with preconditioned Crank-Nicolson proposal and Gibbs sampling scheme is used to estimate the merit parameters. The proposed methodology has been shown to converge, and a ranking-based stopping rule is proposed. We apply this methodology to rank the states and union territories (UTs) of India using data from the National Family Health Survey-5. We estimate and compare rankings under different regimes, e.g., all states/UTs, low-income states/UTs, mid-income states/UTs, and states/UTs by removing high-income states/UTs. Our results reveal meaningful deviations between economic standing and overall performance.
We present the first regret bound for classical online Q-learning in infinite-horizon discounted Markov decision processes (MDPs), without relying on optimism or bonus terms. We first analyze Boltzmann Q-learning with decaying temperature and show that its regret depends critically on the suboptimality gap of the MDP: for sufficiently large gaps, the regret is sublinear, while for small gaps it deteriorates and can approach linear growth. To address this limitation, we study a Smoothed $\epsilon_n$-Greedy exploration scheme that combines $\epsilon_n$-greedy and Boltzmann exploration, for which we prove a gap-robust regret bound of near-$\tilde{O}(N^{9/10})$. We also obtain sample complexity guarantees, with both regret and sample complexity bounds holding with high probability. To analyze these algorithms, we develop a high-probability concentration bound for contractive Markovian stochastic approximation with iterate- and time-dependent transition dynamics. This bound may be of independent interest as the contraction factor in our framework is allowed to converge to one asymptotically.