arXiv subjects

Prashant Shekhar

Publications and source records attributed to Prashant Shekhar.

At least 19 recordsLinked to original sources

arXiv2026

Decision-Calibrated Conformal Uncertainty for Pacing Decisions in Streaming Advertising

We develop a decision-calibrated conformal framework for pacing decisions in streaming advertising. Pacing depends on uncertain future inventory, demand pressure, incremental response, and member-experience load. Instead of calibrating a generic forecast residual, the framework measures forecast error by its largest impact on the policies that could actually be deployed. The main theorem shows that the proposed score is the smallest valid uncertainty measure that uniformly protects all deployable pacing policies. Geometrically, it is the support function of the signed policy sensitivity set. Split conformal calibration gives finite-sample coverage for this score. A high-dimensional separation theorem shows that traditional residual calibration can be arbitrarily more conservative by paying for nuisance inventory dimensions, and a robust pacing result combines inventory, response, and experience uncertainty. On public-data-calibrated pacing replays built from Criteo Uplift and KuaiRand datasets, traditional conformal pacing remains unresolved with high residual radii of 7236.7 on Criteo and 4629.4 on KuaiRand. With the proposed decision calibration approach, the uncertainty radii are reduced to 18.4 and 278.6 respectively, with separate margins for value, delivery, budget, and member load. On Criteo, the proposed method certifies a less aggressive pacing policy than the point-forecast baseline, and reduces held-out any-violation rate from 16.7% to 3.3%, with zero budget and member-load violations. On KuaiRand, the choice remains unresolved. In a nutshell, the paper establishes that forecasts, response estimates, and member-experience models should be judged by whether they shrink the uncertainty that the pacing decision uses, as this leads to confident decisions that are not overly conservative.

stat.ML↗

arXiv2026

Privacy-Robust Incrementality Measurement for Advertising Systems under Signal Loss

Advertising platforms use randomized lift tests to measure incrementality, but privacy-preserving reporting systems degrade the observed signal through match-rate loss, linkability loss, attribution-window loss, aggregation-threshold suppression, randomized reporting noise, and segment-heterogeneous signal loss. This paper formulates privacy-constrained advertising measurement as a robust causal decision problem under the mentioned signal losses. Given a randomized experiment and an ambiguity set for privacy-induced degradation, the framework projects the observation-compatible fiber of clean/unfiltered experimental worlds onto the incrementality functional and returns certified, rejected, and unresolved decisions. The main result gives a sharp decision frontier. Reports outside the frontier support uniformly valid certification or rejection, whereas reports inside it contain too little information for any method to uniformly distinguish above-threshold incrementality from non-incrementality. Supporting results give finite-sample certification, sample-complexity guarantees, a minimax lower bound showing that signal loss reduces effective information, and a reporting-granularity tradeoff. On 2.0M Criteo Uplift rows and the 64K-row Hillstrom email experiment, clean conversion lift is positive in both datasets, with lifts 0.00112 and 0.00495, respectively. Population certification survives mild degradation in Criteo and severe degradation in Hillstrom, while all considered finite-sample stress settings in both datasets remain unresolved after simultaneous uncertainty and reporting noise are included. Overall, the research contributes a decision-theoretic layer for privacy-aware incrementality measurement whose output is the strongest causal-claim justified by degraded ads signals.

stat.ML↗

arXiv2026

PILOT: A Data-Free Continual Learning Approach for Real-Time Semantic Segmentation via Boundary Guidance

Real-time semantic segmentation models offer an excellent balance between accuracy and inference speed. However, deploying these models in dynamic real world environments often requires the ability to learn novel classes incrementally without retraining on the entire dataset. This capability is known as continual learning. In this regard, the standard fine-tuning methods in deep learning often fail due to catastrophic forgetting, where the model learns new information but forgets previously trained and learned classes. Contributing to this crucial domain, the current paper proposes a novel continual learning framework tailored for PIDNet, which is a widely cited state-of-the-art real-time semantic segmentation model. Our method, PILOT(Parallel Incremental Learning Over Time), introduces a real-time and lightweight strategy by implementing a parallel Derivative-branch (D-branch) designed to capture the high frequency boundary information of novel classes while freezing the trained parameters of the original segmentation network. This novel setup allows the model to adapt to new semantic categories while preserving the knowledge of previously learned classes. By using only data associated with the new class, our model significantly reduces training overhead. Experimental results demonstrate that our approach successfully segments new classes while maintaining high mean Intersection over Union (mIoU) on the original base classes, thereby comfortably outperforming all major continual learning approaches in this domain. Overall, PILOT is shown to effectively mitigate catastrophic forgetting with minimal impact on inference latency, thus maintaining real-time performance.

cs.CV↗

arXiv2026

Choosing Online Experiment Designs under Interference in Ads, Recommendations, and Member-Experience Systems

Online experiments in ads, recommendation, and member-experience systems are often planned before the dominant interference mechanism is known. A treatment may propagate through budgets, inventory, producer exposure, graph spillovers, or temporal carryover, making the randomization design itself a statistical decision. We formulate this problem as robust design selection over uncertain exposure mechanisms. Given a finite catalog of six implementable designs, the selector compares each design by worst-case planning risk over an ambiguity set. The risk combines exposure bias, assignment-unit variance, minimum detectable effect, contamination or carryover, operational cost, and estimand mismatch. For theoretical justification, the paper develops a geometry-aware guarantee, stating that design bias is bounded by Wasserstein distance to the launch exposure distribution, and this penalty is minimax tight under Lipschitz exposure response. We also prove finite-catalog approximation and a robust selector theorem with excess-risk control, exact recovery under separation, and certified shortlists when the risk surface is flat. Empirically, the same selector gives different recommendations across samples from public datasets. It selects user-randomization on Criteo ads with dimensionless robust risk 1.295, switchbacks on Open Bandit-bts/men with risk 2.105, and cluster-randomization on KuaiRand with risk 2.240. The Open Bandit case stresses known but uneven logging support, with propensities from 0.00006 to 0.594 and a 5.17% IPS effective-sample share. Overall, the paper contributes an interference-aware experiment design framework based on mechanism-robust design decisions, where the output is either a justified design choice or an uncertainty shortlist.

stat.ML↗

arXiv2026

Support-aware offline policy selection for advertising marketplaces

Logged advertising auctions make offline reserve-price evaluation attractive but risky. Replay tables can identify policies with large apparent yield gains, yet they can also hide weak threshold support, multiple-comparison effects, subgroup harm, and bidder-response uncertainty. Existing replay and off-policy evaluation methods estimate or rank policy values, but they do not directly answer the operational question of whether the available evidence is strong enough to justify validation. This paper develops a support-aware offline decision framework for reserve-policy selection. Rather than outputting a single point-estimate winner, the framework converts logged evidence into a conservative decision object consisting of certified policies, statistically dominated alternatives, and unresolved candidates requiring further validation. The main theoretical result gives a unified finite-catalog guarantee showing that, under simultaneous uncertainty control and conservative support gates, the framework preserves the best gate-passing policy while eliminating only policies with certified regret. Supporting results characterize support-localized replay generalization, establish information-theoretic threshold-resolution limits, and quantify when heterogeneous bidder response can overturn localized replay rankings. Experiments on iPinYou real-time-bidding logs show that the leading reserve rule achieves a 47.66% replay lift in season two, a 40.71% simultaneous lower-bound lift, and a 43.87% frozen out-of-time replay lift in season three. The framework reduces a 19-policy catalog to a two-policy validation shortlist while certifying non-harm across 44 advertiser, exchange, and region segments. The results support the central claim that offline reserve-policy evaluation should produce certified validation decisions rather than point-estimate rankings alone.

stat.ML↗

arXiv2026

Decision Support for Marketplace Policies under Incomplete Evidence: From Replay to Launch Readiness

Marketplace platforms routinely evaluate pricing and allocation policies using logged observational data, yet strong offline performance does not imply that a policy is safe to deploy. In real-time bidding (RTB) marketplaces, reserve-price and floor-policy changes affect not only revenue but also fill, advertiser value, budget pacing, and competition across auctions, creating feedback and interference. The central problem is therefore not to estimate whether a policy improves an offline metric, but to determine whether the available evidence justifies direct launch or only further validation. In this regard, we propose a support-aware decision-support system (DSS) that distinguishes promising from actionable evidence. The framework integrates replay, support-aware off-policy evaluation (OPE), conservative lower-bound ranking, multi-sided guardrails, out-of-time validation, sensitivity analysis, and interference-aware validation design into a claim-preserving pipeline that outputs a launch-readiness classification rather than a single performance estimate. Applying the framework to iPinYou-style RTB logs, we identify a margin-gated floor policy as the leading candidate, with a 47.7% replay yield lift, a 45.8% conservative lower-tail lift, and stable out-of-time performance. However, the framework does not recommend direct launch. A decision-rule ablation shows that simplified pipelines select the same policy but incorrectly recommend deployment, leaving key causal assumptions unresolved. In contrast, the proposed DSS selects the same policy but changes the action to online validation, reflecting missing evidence on propensities, bidder response, and interference. Overall, the contribution is a reproducible DSS protocol that prevents decision overclaim under partial identification and converts offline evaluation into an auditable, action-oriented recommendation.

stat.AP↗

arXiv2025

Adversarial-Resilient RF Fingerprinting: A CNN-GAN Framework for Rogue Transmitter Detection

Radio Frequency Fingerprinting (RFF) has evolved as an effective solution for authenticating devices by leveraging the unique imperfections in hardware components involved in the signal generation process. In this work, we propose a Convolutional Neural Network (CNN) based framework for detecting rogue devices and identifying genuine ones using softmax probability thresholding. We emulate an attack scenario in which adversaries attempt to mimic the RF characteristics of genuine devices by training a Generative Adversarial Network (GAN) using In-phase and Quadrature (IQ) samples from genuine devices. The proposed approach is verified using IQ samples collected from ten different ADALM-PLUTO Software Defined Radios (SDRs), with seven devices considered genuine, two as rogue, and one used for validation to determine the threshold.

cs.CR↗

arXiv2025

Advancing Practical Homomorphic Encryption for Federated Learning: Theoretical Guarantees and Efficiency Optimizations

Federated Learning (FL) enables collaborative model training while preserving data privacy by keeping raw data locally stored on client devices, preventing access from other clients or the central server. However, recent studies reveal that sharing model gradients creates vulnerability to Model Inversion Attacks, particularly Deep Leakage from Gradients (DLG), which reconstructs private training data from shared gradients. While Homomorphic Encryption has been proposed as a promising defense mechanism to protect gradient privacy, fully encrypting all model gradients incurs high computational overhead. Selective encryption approaches aim to balance privacy protection with computational efficiency by encrypting only specific gradient components. However, the existing literature largely overlooks a theoretical exploration of the spectral behavior of encrypted versus unencrypted parameters, relying instead primarily on empirical evaluations. To address this gap, this paper presents a framework for theoretical analysis of the underlying principles of selective encryption as a defense against model inversion attacks. We then provide a comprehensive empirical study that identifies and quantifies the critical factors, such as model complexity, encryption ratios, and exposed gradients, that influence defense effectiveness. Our theoretical framework clarifies the relationship between gradient selection and privacy preservation, while our experimental evaluation demonstrates how these factors shape the robustness of defenses against model inversion attacks. Collectively, these contributions advance the understanding of selective encryption mechanisms and offer principled guidance for designing efficient, scalable, privacy-preserving federated learning systems.

cs.CR↗

arXiv2025

Secure Distributed Learning for CAVs: Defending Against Gradient Leakage with Leveled Homomorphic Encryption

Federated Learning (FL) enables collaborative model training across distributed clients without sharing raw data, making it a promising approach for privacy-preserving machine learning in domains like Connected and Autonomous Vehicles (CAVs). However, recent studies have shown that exchanged model gradients remain susceptible to inference attacks such as Deep Leakage from Gradients (DLG), which can reconstruct private training data. While existing defenses like Differential Privacy (DP) and Secure Multi-Party Computation (SMPC) offer protection, they often compromise model accuracy. To that end, Homomorphic Encryption (HE) offers a promising alternative by enabling lossless computation directly on encrypted data, thereby preserving both privacy and model utility. However, HE introduces significant computational and communication overhead, which can hinder its practical adoption. To address this, we systematically evaluate various leveled HE schemes to identify the most suitable for FL in resource-constrained environments due to its ability to support fixed-depth computations without requiring costly bootstrapping. Our contributions in this paper include a comprehensive evaluation of HE schemes for real-world FL applications, a selective encryption strategy that targets only the most sensitive gradients to minimize computational overhead, and the development of a full HE-based FL pipeline that effectively mitigates DLG attacks while preserving model accuracy. We open-source our implementation to encourage reproducibility and facilitate adoption in safety-critical domains.

cs.CR↗

arXiv2025

Cross-Model Transferability of Adversarial Patches in Real-time Segmentation for Autonomous Driving

Adversarial attacks pose a significant threat to deep learning models, particularly in safety-critical applications like healthcare and autonomous driving. Recently, patch based attacks have demonstrated effectiveness in real-time inference scenarios owing to their 'drag and drop' nature. Following this idea for Semantic Segmentation (SS), here we propose a novel Expectation Over Transformation (EOT) based adversarial patch attack that is more realistic for autonomous vehicles. To effectively train this attack we also propose a 'simplified' loss function that is easy to analyze and implement. Using this attack as our basis, we investigate whether adversarial patches once optimized on a specific SS model, can fool other models or architectures. We conduct a comprehensive cross-model transferability analysis of adversarial patches trained on SOTA Convolutional Neural Network (CNN) models such PIDNet-S, PIDNet-M and PIDNet-L, among others. Additionally, we also include the Segformer model to study transferability to Vision Transformers (ViTs). All of our analysis is conducted on the widely used Cityscapes dataset. Our study reveals key insights into how model architectures (CNN vs CNN or CNN vs. Transformer-based) influence attack susceptibility. In particular, we conclude that although the transferability (effectiveness) of attacks on unseen images of any dimension is really high, the attacks trained against one particular model are minimally effective on other models. And this was found to be true for both ViT and CNN based models. Additionally our results also indicate that for CNN-based models, the repercussions of patch attacks are local, unlike ViTs. Per-class analysis reveals that simple-classes like 'sky' suffer less misclassification than others. The code for the project is available at: https://github.com/p-shekhar/adversarial-patch-transferability

cs.CV↗

arXiv2022

Machine learning based surrogate modeling with SVD enabled training for nonlinear civil structures subject to dynamic loading

The computationally expensive estimation of engineering demand parameters (EDPs) via finite element (FE) models, while considering earthquake and parameter uncertainty limits the use of the Performance Based Earthquake Engineering framework. Attempts have been made to substitute FE models with surrogate models, however, most of these models are a function of building parameters only. This necessitates re-training for earthquakes not previously seen by the surrogate. In this paper, the authors propose a machine learning based surrogate model framework, which considers both these uncertainties in order to predict for unseen earthquakes. Accordingly,earthquakes are characterized by their projections on an orthonormal basis, computed using SVD of a representative ground motion suite. This enables one to generate large varieties of earthquakes by randomly sampling these weights and multiplying them with the basis. The weights along with the constitutive parameters serve as inputs to a machine learning model with EDPs as the desired output. Four competing machine learning models were tested and it was observed that a deep neural network (DNN) gave the most accurate prediction. The framework is validated by using it to successfully predict the peak response of one-story and three-story buildings represented using stick models, subjected to unseen far-field ground motions.

stat.ML↗

arXiv2021

A Forward Backward Greedy approach for Sparse Multiscale Learning

Multiscale Models are known to be successful in uncovering and analyzing the structures in data at different resolutions. In the current work we propose a feature driven Reproducing Kernel Hilbert space (RKHS), for which the associated kernel has a weighted multiscale structure. For generating approximations in this space, we provide a practical forward-backward algorithm that is shown to greedily construct a set of basis functions having a multiscale structure, while also creating sparse representations from the given data set, making representations and predictions very efficient. We provide a detailed analysis of the algorithm including recommendations for selecting algorithmic hyper-parameters and estimating probabilistic rates of convergence at individual scales. Then we extend this analysis to multiscale setting, studying the effects of finite scale truncation and quality of solution in the inherent RKHS. In the last section, we analyze the performance of the approach on a variety of simulation and real data sets, thereby justifying the efficiency claims in terms of model quality and data reduction.

cs.LG↗

arXiv2020

ALPS: A Unified Framework for Modeling Time Series of Land Ice Changes

Modeling time series is a research focus in cryospheric sciences because of the complexity and multiscale nature of events of interest. Highly non-uniform sampling of measurements from different sensors with different levels of accuracy, as is typical for measurements of ice sheet elevations, makes the problem even more challenging. In this paper, we propose a spline-based approximation framework (ALPS - Approximation by Localized Penalized Splines) for modeling time series of land ice changes. The localized support of the B-spline basis functions enable robustness to non-uniform sampling, a considerable improvement over other global and piecewise local models. With features like, discrete-coordinate-difference-based penalization and two-level outlier detection, ALPS further guarantees the stability and quality of approximations. ALPS incorporates rigorous model uncertainty estimates with all approximations. As demonstrated by examples, ALPS performs well for a variety of data sets, including time series of ice sheet thickness, elevation, velocity, and terminus locations. The robust estimation of time series and their derivatives facilitates new applications, such as the reconstruction of high-resolution elevation change records by fusing sparsely sampled time series of ice sheet thickness changes with modeled firn thickness changes, and the analysis of the relationship between different outlet glacier observations to gain new insight into processes and forcing.

stat.AP↗

arXiv2020

Hierarchical regularization networks for sparsification based learning on noisy datasets

We propose a hierarchical learning strategy aimed at generating sparse representations and associated models for large noisy datasets. The hierarchy follows from approximation spaces identified at successively finer scales. For promoting model generalization at each scale, we also introduce a novel, projection based penalty operator across multiple dimension, using permutation operators for incorporating proximity and ordering information. The paper presents a detailed analysis of approximation properties in the reconstruction Reproducing Kernel Hilbert Spaces (RKHS) with emphasis on optimality and consistency of predictions and behavior of error functionals associated with the produced sparse representations. Results show the performance of the approach as a data reduction and modeling strategy on both synthetic (univariate and multivariate) and real datasets (time series). The sparse model for the test datasets, generated by the presented approach, is also shown to efficiently reconstruct the underlying process and preserve generalizability.

cs.LG↗

arXiv2019

Hierarchical Data Reduction and Learning

This paper describes a hierarchical learning strategy for generating sparse representations of multivariate datasets. The hierarchy arises from approximation spaces considered at successively finer scales. A detailed analysis of stability, convergence and behavior of error functionals associated with the approximations are presented, along with a well chosen set of applications. Results show the performance of the approach as a data reduction mechanism for both synthetic (univariate and multivariate) and real datasets (geospatial and numerical model outcomes). The sparse representation generated is shown to efficiently reconstruct data and minimize error in prediction.

stat.ML↗

arXiv2018

Controlling evanescent waves using silicon photonic all-dielectric metamaterials for dense integration

Ultra-compact, densely integrated optical components manufactured on a CMOS-foundry platform are highly desirable for optical information processing and electronic-photonic co-integration. However, the large spatial extent of evanescent waves arising from nanoscale confinement, ubiquitous in silicon photonic devices, causes significant cross-talk and scattering loss. Here, we demonstrate that anisotropic all-dielectric metamaterials open a new degree of freedom in total internal reflection to shorten the decay length of evanescent waves. We experimentally show the reduction of cross-talk by greater than 30 times and the bending loss by greater than 3 times in densely integrated, ultra-compact photonic circuit blocks. Our prototype all-dielectric metamaterial-waveguide achieves a low propagation loss of approximately 3.7 dB/cm, comparable to those of silicon strip waveguides. Our approach marks a departure from interference-based confinement as in photonic crystals or slot waveguides, which utilize nanoscale field enhancement. Its ability to suppress evanescent waves without substantially increasing the propagation loss shall pave the way for all-dielectric metamaterial-based dense integration.

physics.optics↗

arXiv2017

Momentum-resolved Electron Energy Loss Spectroscopy for Mapping the Photonic Density of States

Strong nanoscale light-matter interaction is often accompanied by ultra-confined photonic modes and large momentum polaritons existing far beyond the light cone. A direct probe of such phenomena is difficult due to the momentum mismatch of these modes with free space light however, fast electron probes can reveal the fundamental quantum and spatially dispersive behavior of these excitations. Here, we use momentum-resolved electron energy loss spectroscopy (q-EELS) in a transmission electron microscope to explore the optical response of plasmonic thin films including momentum transfer up to wavevectors (q) significantly exceeding the light line wave vector. We show close agreement between experimental q-EELS maps, theoretical simulations of fast electrons passing through thin films and the momentum-resolved photonic density of states (q-PDOS) dispersion. Although a direct link between q-EELS and the q-PDOS exists for an infinite medium, here we show fundamental differences between q-EELS measurements and the q-PDOS that must be taken into consideration for realistic finite structures with no translational invariance along the direction of electron motion. Our work paves the way for using q-EELS as the preeminent tool for mapping the q-PDOS of exotic phenomena with large momenta (high-q) such as hyperbolic polaritons and spatially-dispersive plasmons.

physics.optics↗

arXiv2014

Strong Coupling in Hyperbolic Metamaterials

Nanoscale light-matter interaction in the weak-coupling regime has been achieved with unique hyperbolic metamaterial modes possessing a high density of states. Here, we show strong coupling between intersubband transitions (ISBTs) of a multiple quantum well (MQW) slab and the bulk polariton modes of a hyperbolic metamaterial (HMM). These HMM modes have large wave vectors (high-k modes) and are normally evanescent in conventional materials. We analyze a metal-dielectric practical multilayer HMM structure consisting of a highly doped semiconductor acting as a metallic layer and an active multiple quantum well dielectric slab. We observe delocalized metamaterial mode interaction with the active materials distributed throughout the structure. Strong coupling and characteristic anticrossing with a maximum Rabi splitting (RS) energy of up to 52 meV is predicted between the high-k mode of the HMMand the ISBT, a value approximately 10.5 times greater than the ISBT linewidth and 4.5 times greater than the material loss of the structure. The scalability and tunability of the RS energy in an active semiconductor metamaterial device have potential applications in quantum well infrared photodetectors and intersubband light-emitting devices.

physics.optics↗