Chirag Agarwal

• Postdoctoral Research Fellow
• PostDoc Position at Harvard University

Large language models have emerged as powerful tools for general intelligence, showcasing advanced natural language processing capabilities that find applications across diverse domains. Despite their impressive performance, recent studies have highlighted the potential for significant enhancements in LLMs' task-specific performance through fine-tu...

As machine learning models are increasingly being employed in various high-stakes settings, it becomes important to ensure that predictions of these models are not only adversarially robust, but also readily explainable to relevant stakeholders. However, it is unclear if these two notions can be simultaneously achieved or if there exist trade-offs...

As explanations are increasingly used to understand the behavior of graph neural networks (GNNs), evaluating the quality and reliability of GNN explanations is crucial. However, assessing the quality of GNN explanations is challenging as existing graph datasets have no or unreliable ground-truth explanations. Here, we introduce a synthetic graph da...

Graph unlearning, which involves deleting graph elements such as nodes, node labels, and relationships from a trained graph neural network (GNN) model, is crucial for real-world applications where data elements may become irrelevant, inaccurate, or privacy-sensitive. However, existing methods for graph unlearning either deteriorate model weights sh...

As post hoc explanations are increasingly used to understand the behavior of graph neural networks (GNNs), it becomes crucial to evaluate the quality and reliability of GNN explanations. However, assessing the quality of GNN explanations is challenging as existing graph datasets have no or unreliable ground-truth explanations for a given task. Here...

Background and methodology Electrocardiogram (ECG) deep learning (DL) has promise to improve the outcomes of patients with cardiovascular abnormalities. In ECG DL, researchers often use convolutional neural networks (CNNs) and traditionally use the full duration of raw ECG waveforms that create redundancies in feature learning and result in inaccur...

While several types of post hoc explanation methods (e.g., feature attribution methods) have been proposed in recent literature, there is little to no work on systematically benchmarking these methods in an efficient and transparent manner. Here, we introduce OpenXAI, a comprehensive and extensible open source framework for evaluating and benchmark...

Purpose Persistent sustained attention deficit (SAD) after continuous positive airway pressure (CPAP) treatment is a source of quality of life and occupational impairment in obstructive sleep apnea (OSA). However, persistent SAD is difficult to predict in patients initiated on CPAP treatment. We performed secondary analyses of brain magnetic resona...

As attribution-based explanation methods are increasingly used to establish model trustworthiness in high-stakes situations, it is critical to ensure that these explanations are stable, e.g., robust to infinitesimal perturbations to an input. However, previous works have shown that state-of-the-art explanation methods generate unstable explanations...

As machine learning models are increasingly employed to assist human decision-makers, it becomes critical to communicate the uncertainty associated with these model predictions. However, the majority of work on uncertainty has focused on traditional probabilistic or ranking approaches - where the model assigns low probabilities or scores to uncerta...

Counterfactual explanations and adversarial examples have emerged as critical research areas for addressing the explainability and robustness goals of machine learning (ML). While counterfactual explanations were developed with the goal of providing recourse to individuals adversely impacted by algorithmic decisions, adversarial examples were desig...

As Graph Neural Networks (GNNs) are increasingly employed in real-world applications, it becomes critical to ensure that the stakeholders understand the rationale behind their predictions. While several GNN explanation methods have been proposed recently, there has been little to no work on theoretically analyzing the behavior of these methods or s...

Perturbation-based explanation methods often measure the contribution of an input feature to an image classifier’s outputs by heuristically removing it via e.g. blurring, adding noise, or graying out, which often produce unrealistic, out-of-samples. Instead, we propose to integrate a generative inpainter into three representative attribution method...

As the representations output by Graph Neural Networks (GNNs) are increasingly employed in real-world applications, it becomes important to ensure that these representations are fair and stable. In this work, we establish a key connection between counterfactual fairness and stability and leverage it to propose a novel framework, NIFTY (uNIfying Fai...

As machine learning black boxes are increasingly being deployed in critical domains such as healthcare and criminal justice, there has been a growing emphasis on developing techniques for explaining these black boxes in a post hoc manner. In this work, we analyze two popular post hoc interpretation techniques: SmoothGrad which is a gradient based m...

In machine learning, a question of great interest is understanding what examples are challenging for a model to classify. Identifying atypical examples helps inform safe deployment of models, isolates examples that require further human inspection, and provides interpretability into model behavior. In this work, we propose Variance of Gradients (VO...

Adversarial training has been the topic of dozens of studies and a leading method for defending against adversarial attacks. Yet, it remains unknown (a) how adversarially-trained classifiers (a.k.a "robust" classifiers) generalize to new types of out-of-distribution examples; and (b) what hidden representations were learned by robust networks. In t...

Study objectives: Nocturnal blood pressure (BP) profile shows characteristic abnormalities in obstructive sleep apnea (OSA), namely acute post-apnea BP surges and non-dipping BP. These abnormal BP profiles provide prognostic clues indicating increased cardiovascular disease (CVD) risk. We developed a deep neural network model to perform computeriz...

In late 2019, a new Coronavirus disease, referred to as Corona virus disease 2019 (COVID-19), emerged in Wuhan city, Hubei, China, and resulted in a global pandemic---claiming a large number of lives and affecting billions all around the world. The current global standard used in diagnosis of COVID-19 in suspected cases is the real-time polymerase...

Attribution methods can provide powerful insights into the reasons for a classifier's decision. We argue that a key desideratum of an explanation method is its robustness to input hyperparameters which are often randomly set or empirically tuned. High sensitivity to arbitrary hyperparameter choices does not only impede reproducibility but also ques...

Attribution methods can provide powerful insights into the reasons for a classifier's decision. We argue that a key desideratum of an explanation method is its robustness to input hyperparameters which are often randomly set or empirically tuned. High sensitivity to arbitrary hyperparameter choices does not only impede reproducibility but also ques...

The lack of interpretability in current deep learning models causes serious concerns as they are extensively used for various life-critical applications. Hence, it is of paramount importance to develop interpretable deep learning models. In this paper, we consider the problem of blind deconvolution and propose a novel model-aware deep architecture...

Interpretability methods often measure the contribution of an input feature to an image classifier's decisions by heuristically removing it via e.g. blurring, adding noise, or graying out, which often produce unrealistic, out-of-samples. Instead, we propose to integrate a generative inpainter into three representative attribution methods to remove...

Interpretability methods often measure the contribution of an input feature to an image classifier's decisions by heuristically removing it via e.g. blurring, adding noise, or graying out, which often produce unrealistic, out-of-samples. Instead, we propose to integrate a generative inpainter into three representative attribution methods to remove...

Deep neural networks (DNNs) have achieved state-of-the-art results in various pattern recognition tasks. However, they perform poorly on out-of-distribution adversarial examples i.e. inputs that are specifically crafted by an adversary to cause DNNs to misbehave, questioning the security and reliability of applications. In this paper, we encourage...

The location and type of adipose tissue is an important factor in metabolic syndrome. A database of picture archiving and communication system (PACS) derived abdominal computerized tomography (CT) images from a large health care provider, Geisinger, was used for large-scale research of the relationship of volume of subcutaneous adipose tissue (SAT)...

Deep Neural Networks(DNN) have excessively advanced the field of computer vision by achieving state of the art performance in various vision tasks. These results are not limited to the field of vision but can also be seen in speech recognition and machine translation tasks. Recently, DNNs are found to poorly fail when tested with samples that are c...

Study of connectivity of neural circuits is an essential step towards a better understanding of functioning of the nervous system. With the recent improvement in imaging techniques, high-resolution and high-volume images are being generated requiring automated segmentation techniques. We present a pixel-wise classification method based on Bayesian...

We propose a novel architecture based on the strucuture of AutoEncoders. The paper introduces CrossEncoders - an AutoEncoder architecture which uses cross-connections to connect layers (both adjacent and non-adjacent) in the encoder and decoder side of the network respectively. The network incorporates both global and local information in the lower...

This paper presents a novel approach to increase the performance bounds of image steganography under the criteria of minimizing distortion. The proposed approach utilizes a steganalysis convolutional neural network (CNN) framework to understand an image's model and embed in less detectable regions to preserve the model. In other word, the trained s...

This paper presents two novel approaches to increase performance bounds of image steganography under the criteria of minimizing distortion. First, in order to efficiently use the images' capacities, we propose using parallel images in the embedding stage. The result is then used to prove sub-optimality of the message distribution technique used by...

We propose a novel neural network structure called CrossNets, which considers architectures on directed acyclic graphs. This structure builds on previous generalizations of feed forward models, such as ResNets, by allowing for all forward cross connections between layers (both adjacent and non-adjacent). The addition of cross connections among the...

Visual saliency signifies the region of perceptual significance in an image to Human Visual System (HVS).It is the inherent quality that discriminates an object from its neighbors. The Saliency map computes such regions in an image. In this paper, we present the application of Wavelet domain data hiding in such prominent regions. The proposed algor...

Visual saliency signifies the region of perceptual significance in an image to human vision system (HVS). It is the inherent quality that discriminates an object from its neighbors. The Saliency map computes such regions in an image. In this paper, we present the application of Wavelet domain data hiding in such prominent regions. The proposed algo...

In recent years the study of watermarking using CDMA technique in transform domain is a widely popular method. We incorporated modification in existing algorithm with improved imperceptibility of embedded image while making it capable to extract the hidden image by using correlation technique. The original cover image is decomposed in two stages, f...