We present TenderWallet, a CBDC system that offers a balance of inclusive privacy-preserving access and complete regulatory compliance. Our proposed system provides a CBDC functionality that is responsive to user behaviour, and in case of non-compliance by the user while availing the CBDC service, compensates the indirect monetary loss induced to the nation state, through collateralization. The TenderWallet system is also hybrid, such that it dynamically works as one variant of two well established offline CBDC models, thereby reducing the computational overhead of the service. We detail the behaviour of the TenderWallet system, analyze its deployment characteristics, and argue that it is superior to other potential hybrid CBDC designs. We conclude by showing that TenderWallet is near optimal, while considering compensation to the state and low-compute privacy and transparency of users as the main requirements of a CBDC system.

The marked shift of application developers to serverless computing has led to an increase in the number of cyberattacks and privacy concerns, thus prompting the need for secure serverless workflows. We propose TrapShield, a honeypots-based, secure and privacy preserving framework to protect serverless computing applications from insider and outsider attacks. It utilizes honeypots to deceive the attackers and penalize them by redirecting to a random set of dummy functions forming a cycle. Evaluations on Google Cloud Platform and Amazon Web Services for three popular serverless applications show TrapShield’s effectiveness in reducing costs for thwarting attacks while maintaining high runtime performance (approximately 1.3 seconds for an airline booking application).

Several nations across the world are contemplating optimal design choices for Central Bank Digital Currencies (CBDCs). We present LimboCoin, an analytical framework for an arbitrary token based CBDC protocol. LimboCoin, under practical state-of-the-art assumptions on secure system design and protocol incentivization, considers an adversarial behaviour to achieve denial-of-service on the CBDC’s associated system and token economy. LimboCoin outlines the quality of the CBDC system and economy resultant from the interaction of honest and adversarial users in the CBDC’s jurisdiction. Through LimboCoin, we show that in the worst case, the number of compromised CBDC wallets in operation can exceed the number of legitimate wallets in operation, within the CBDC’s jurisdiction. We also show that in the worst case, the value associated with victim token transactions that are denied service exceeds 90 percent of the value associated with token transactions that are in legitimate service.

The recent advancement of serverless computing in the widespread deployment of applications prompts the need to protect serverless workflows against cloud vulnerabilities and threats. We propose PrivFlow, a workflow-centric, privacy preserving framework to protect the information flow in serverless computing applications in semi-honest (S-PrivFlow) and malicious (M-PrivFlow) adversarial settings. An Authenticated Data Structure is used to store the valid workflows encoded in the proposed format. The validation of workflows is performed in a privacy preserving manner that leaks no sensitive information to any unauthorized user. We focus on the two most prevalent attacks on the serverless cloud platforms, namely the Denial-of-Wallet and Wrong Function Invocation attacks. We demonstrate that PrivFlow mitigates both of these attacks. Further, we evaluate PrivFlow on the popular benchmark application- Hello Retail, and a customized scaled application. Though the comparison with the state-of-the-art approaches in terms of the runtime performance shows a latency of 1.6 times for S-PrivFlow and 8 times for M-PrivFlow, the PrivFlow provides high security and privacy. PrivFlow acts as a wrapper to the application resulting in no change to the source code.

Scene graph captures rich semantic information of an image by representing objects and their relationships as nodes and edges of a graph. Recent works have demonstrated that scene graph representation improves the performance of various computer vision tasks such as image retrieval, action recognition, visual question answering. Computationally efficient scene graph generation methods are required to leverage scene graphs in various real-world applications (e.g., autonomous driving, robotics). A typical scene graph generation model consists of two modules: (i) object detector and (ii) scene graph classifier. The scene graph classifier module predicts the object category and object-object relationships. The presence of a quadratic number of potential edges poses a major challenge in the scene graph classification task. Detecting the relationship between each object pair using the traditional approach is computationally intensive and non-scalable. To address this issue, we propose a novel module named EdgeNet that directly predicts the set of relevant edges and helps to prune out a significant number of unrelated object pairs, thereby improving the effectiveness and efficiency of the scene graph classifier. The proposed EdgeNet is a generic module and can be plugged into an existing scene graph classifier. Experimental results highlight the effectiveness and efficiency of the proposed approach on the Visual Genome dataset.

Fraudulent transaction detection is a difficult problem for an individual bank, since the number of fraudulent transactions within a single bank’s records is significantly less compared to the day-to-day regular transactions it processes. Hence, due to this extreme data imbalance, training a classifier is difficult. Also, the model will not be able to learn from different types of fraudulent transactions, which a single bank’s database lacks. Collaboration between banks is the only way to achieve a generalized model, but banks will not share their data with each other due to competition and regulatory restrictions. Federated Learning can be leveraged here to solve this problem. However, in a cross-silo setting like this, the data held by different banks will be different in terms of distribution and hence follows a non-IID scenario across the participants’ datasets. Moreover, we are considering that a minority of the banks could be malicious and will try to disrupt this federated learning process. Hence the problem is to perform federated learning in a non-IID setting with active adversaries involved, which is a new research area under fraud detection. We perform non-IID partitioning of the transaction dataset to simulate 10 banks or silos. Then, for benchmark, we perform federated averaging with a subset of the banks set as malicious. Furthermore, we propose a novel algorithm - Epsilon Cluster Selection, a filter-based aggregation technique to recognize and prevent malicious nodes from contributing to the global model being trained. We apply this algorithm to the same setting with malicious banks and compare the results.

The rapid proliferation of biometrics has led to growing concerns about the security and privacy of the biometric data (template). A biometric uniquely identifies an individual and unlike passwords, it cannot be revoked or replaced since it is unique and fixed for every individual. To address this problem, many biometric template protection methods using fully homomorphic encryption have been proposed. But, most of them (i) are computationally expensive and practically infeasible (ii) do not support operations over real valued biometric feature vectors without quantization (iii) do not support packing of real valued feature vectors into a ciphertext (iv) require multishot enrollment of users for improved matching performance. To address these limitations, we propose a secure and privacy preserving method for biometric template protection using fully homomorphic encryption. The proposed method is computationally efficient and practically feasible, supports operations over real valued feature vectors without quantization and supports packing of real valued feature vectors into a single ciphertext. In addition, the proposed method enrolls the users using one-shot enrollment. To evaluate the proposed method, we use three face datasets namely LFW, FEI and Georgia tech face dataset. The encrypted face template (for 128 dimensional feature vector) requires 32.8 KB of memory space and it takes 2.83 milliseconds to match a pair of encrypted templates. The proposed method improves the matching performance by ~ 3% when compared to state-of-the-art, while providing high template security.

Demand for computational grid is growing ever since the rise of social networks, Artificial Intelligence based systems (use of machine learning), scientific application, etc. Service providers are aiming towards achieving maximum grid utilization so that they can serve more customers efficiently. The key enablers to achieve optimal grid utilization and better turnaround time is by efficient scheduling of tasks on computational grids. However, designing efficient grid scheduling algorithms is still a challenge due to its complexity (NP-complete). Hence, several near optimal approximation algorithms are designed based on plethora of techniques such as heuristics, bio inspired, genetic, greedy approaches, etc. Thus there is a scope for further improvement in scheduling algorithm to achieve early task completion and better grid utilization for precedence constrained tasks. Moreover, nowadays with advancement in computing hardware there is a preference for parallel task execution strategy rather than sequential task execution. With this there is also a notion of partial dependency between parallel tasks. Thus, there is a need to revisit designing grid scheduling algorithm. Hence, In this paper we propose a novel grid scheduling algorithm for interdependent parallel tasks with varying dependencies (full, partial, no) on computational grid using a greedy approach. Further the correctness and performance of the proposed scheduling algorithm is evaluated by comparing it with proposed brute force scheduling algorithm.