arXiv subjects

Aditi Maheshwari

Publications and source records attributed to Aditi Maheshwari.

4 recordsLinked to original sources

arXiv2026

Let the Neurons Die: Exploiting ReLU-Induced Model Degradation

Rectified linear unit (ReLU) networks can suffer from dying neurons, where units with persistently negative pre-activations produce zero outputs, blocking gradients through their activations. To exploit this failure mode, we present three training-time availability attacks based on data ordering and poisoning. We begin with the basic dynamic data-ordering attack (DOA), which greedily constructs a training prefix by selecting the next example that minimizes the target layer's post-update weight sum, aiming to push ReLU units toward negative pre-activations without modifying training samples or labels. We then develop two poisoning attacks, IG-DOA and IG-SKA, which use gradient inversion to synthesize class-conditioned samples by matching reference gradients in adverse model states constructed through data ordering or soft knockout, respectively. Soft knockout rearranges weights across adjacent layers to concentrate negative contributions. On a fully connected ReLU network trained on MNIST, ordering 100 of 60,000 training examples reduces test accuracy from 96% to 95% after only five epochs. Adding 200 poisoned samples from a single class reduces test accuracy to approximately 86-88% after five epochs in most evaluated conditions, compared with approximately 96% under clean training. These results demonstrate that ReLU-targeted data ordering and poisoning can impair learning without directly modifying the victim model's parameters.

cs.LG↗

arXiv2025

Adversarial Attacks on Deep Learning-Based False Data Injection Detection in Differential Relays

The application of Deep Learning-based Schemes (DLSs) for detecting False Data Injection Attacks (FDIAs) in smart grids has attracted significant attention. This paper demonstrates that adversarial attacks, carefully crafted FDIAs, can evade existing DLSs used for FDIA detection in Line Current Differential Relays (LCDRs). We propose a novel adversarial attack framework, utilizing the Fast Gradient Sign Method, which exploits DLS vulnerabilities by introducing small perturbations to LCDR remote measurements, leading to misclassification of the FDIA as a legitimate fault while also triggering the LCDR to trip. We evaluate the robustness of multiple deep learning models, including multi-layer perceptrons, convolutional neural networks, long short-term memory networks, and residual networks, under adversarial conditions. Our experimental results demonstrate that while these models perform well, they exhibit high degrees of vulnerability to adversarial attacks. For some models, the adversarial attack success rate exceeds 99.7%. To address this threat, we introduce adversarial training as a proactive defense mechanism, significantly enhancing the models' ability to withstand adversarial FDIAs without compromising fault detection accuracy. Our results highlight the significant threat posed by adversarial attacks to DLS-based FDIA detection, underscore the necessity for robust cybersecurity measures in smart grids, and demonstrate the effectiveness of adversarial training in enhancing model robustness against adversarial FDIAs.

cs.LG↗

arXiv2025

ExoFabric: A Re-moldable Textile System for Creating Customizable Soft Goods and Wearable Applications

Fabric has been a fundamental part of human life for thousands of years, providing comfort, protection, and aesthetic expression. While modern advancements have enhanced fabric's functionality, it remains static and unchangeable, failing to adapt to our evolving body shapes and preferences. This lack of adaptability can lead to unsustainable practices, as consumers often buy more items to meet their changing needs. In this paper, we propose ExoFabric, a re-moldable fabric system for customized soft goods applications. We created ExoFabric by embedding thermoplastic threads into fabric through computerized embroidery to allow for tunability between rigid plastic and conformable fabric. We defined a library of design primitives to enable geometric formability, stiffness, and stretchability by identifying suitable fabrics, threads, embroidery parameters, and machine limitations. To facilitate practical applications, we demonstrated practical methods for linking parameters to application requirements, showcasing form-fitting wearables, structural support, and shape-changeable furniture for repeatable or one-time customization.

cs.ET↗

arXiv2023

Guttation Monitor: Wearable Guttation Sensor for Plant Condition Monitoring and Diagnosis

Plant life plays a critical role in the ecosystem. However, it is difficult for humans to perceive plants' reactions because the biopotential and biochemical responses are invisible to humans. Guttation droplets contain various chemicals which can reflect plant physiology and environmental conditions in real-time. Traditionally, these droplets are collected manually and analyzed in the lab with expensive instruments. Here, we introduce the Guttation Monitor, an on-site and low-cost monitoring technology for guttation droplets. It consists of three parts 1) a paper-based microfluidic chip that can collect guttation droplets and perform colorimetric detection of six chemicals, 2) a self-contained and solar-powered camera module that can capture the result from the chip, and 3) an end-user app that can interpret the result. We discuss this technology's design and implementation, conduct evaluations on tomato plants, conduct interviews, and envision how such a technology could enhance the human-plant relationship in four dimensions.

cs.HC↗