Anirban Sarkar

Anindya Sarkar, Anirban Sarkar, Sowrya Gali et al.

Current SOTA adversarially robust models are mostly based on adversarial training (AT) and differ only by some regularizers either at inner maximization or outer minimization steps. Being repetitive in nature during the inner maximization step, they take a huge time to train. We propose a non-iterative method that enforces the following ideas during training. Attribution maps are more aligned to the actual object in the image for adversarially robust models compared to naturally trained models. Also, the allowed set of pixels to perturb an image (that changes model decision) should be restricted to the object pixels only, which reduces the attack strength by limiting the attack space. Our method achieves significant performance gains with a little extra effort (10-20%) over existing AT models and outperforms all other methods in terms of adversarial as well as natural accuracy. We have performed extensive experimentation with CIFAR-10, CIFAR-100, and TinyImageNet datasets and reported results against many popular strong adversarial attacks to prove the effectiveness of our method.

Aditya Chattopadhyay, Anirban Sarkar, Prantik Howlader et al.

Over the last decade, Convolutional Neural Network (CNN) models have been highly successful in solving complex vision problems. However, these deep models are perceived as "black box" methods considering the lack of understanding of their internal functioning. There has been a significant recent interest in developing explainable deep learning models, and this paper is an effort in this direction. Building on a recently proposed method called Grad-CAM, we propose a generalized method called Grad-CAM++ that can provide better visual explanations of CNN model predictions, in terms of better object localization as well as explaining occurrences of multiple object instances in a single image, when compared to state-of-the-art. We provide a mathematical derivation for the proposed method, which uses a weighted combination of the positive partial derivatives of the last convolutional layer feature maps with respect to a specific class score as weights to generate a visual explanation for the corresponding class label. Our extensive experiments and evaluations, both subjective and objective, on standard datasets showed that Grad-CAM++ provides promising human-interpretable visual explanations for a given CNN architecture across multiple tasks including classification, image caption generation and 3D action recognition; as well as in new settings such as knowledge distillation.

Somesh Kumar Prajapati, Suvamoy Changder, Anirban Sarkar

Software as a service or (SaaS) is a new software development and deployment paradigm over the cloud and offers Information Technology services dynamically as "on-demand" basis over the internet. Trust is one of the fundamental security concepts on storing and delivering such services. In general, trust factors are integrated into such existent security frameworks in order to add a security level to entities collaborations through the trust relationship. However, deploying trust factor in the secured cloud environment are more complex engineering task due to the existence of heterogeneous types of service providers and consumers. In this paper, a formal trust management model has been introduced to manage the trust and its properties for SaaS in cloud computing environment. The model is capable to represent the direct trust, recommended trust, reputation etc. formally. For the analysis of the trust properties in the cloud environment, the proposed approach estimates the trust value and uncertainty of each peer by computing decay function, number of positive interactions, reputation factor and satisfaction level for the collected information.