Recent developments in self-supervised learning (SSL) have made it possible to learn data representations without the need for annotations. Inspired by the non-contrastive SSL approach (SimSiam), we introduce a novel framework SimSAM to compute the Semantic Affinity Matrix, which is significant for unsupervised image segmentation. Given an image, SimSAM first extracts features using pre-trained DINO-ViT, then projects the features to predict the correlations of dense features in a non-contrastive way. We show applications of the Semantic Affinity Matrix in object segmentation and semantic segmentation tasks. Our code is available at https://github.com/chandagrover/SimSAM.

BharatSim is an open-source agent-based modelling framework for the Indian population. It can simulate populations at multiple scales, from small communities to states. BharatSim uses a synthetic population created by applying statistical methods and machine learning algorithms to survey data from multiple sources, including the Census of India, the India Human Development Survey, the National Sample Survey, and the Gridded Population of the World. This synthetic population defines individual agents with multiple attributes, among them age, gender, home and work locations, pre-existing health conditions, and socio-economic and employment status. BharatSim's domain-specific language provides a framework for the simulation of diverse models. Its computational core, coded in Scala, supports simulations of a large number of individual agents, up to 50 million. Here, we describe the design and implementation of BharatSim, using it to address three questions motivated by the COVID-19 pandemic in India: (i) When can schools be safely reopened given specified levels of hybrid immunity?, (ii) How do new variants alter disease dynamics in the background of prior infections and vaccinations? and (iii) How can the effects of varied non-pharmaceutical interventions (NPIs) be quantified for a model Indian city? Through its India-specific synthetic population, BharatSim allows disease modellers to address questions unique to this country. It should also find use in the computational social sciences, potentially providing new insights into emergent patterns in social behaviour.

CLIPStyler demonstrated image style transfer with realistic textures using only a style text description (instead of requiring a reference style image). However, the ground semantics of objects in the style transfer output is lost due to style spill-over on salient and background objects (content mismatch) or over-stylization. To solve this, we propose Semantic CLIP-Styler (Sem-CS), that performs semantic style transfer.Sem-CS first segments the content image into salient and non-salient objects and then transfers artistic style based on a given style text description. The semantic style transfer is achieved using global foreground loss (for salient objects) and global background loss (for non-salient objects). Our empirical results, including DISTS, NIMA and user study scores, show that our proposed framework yields superior qualitative and quantitative performance. Our code is available at github.com/chandagrover/sem-cs.

Overcoming the threat of coercion attacks in a cryptographic system has been a top priority for system designers since the birth of cyber-security. One way to overcome such a threat is to leverage implicit memory to construct a defense against rubber-hose attacks where the users themselves do not possess conscious knowledge of the trained password. We propose NeuroCrypt, a coercion-resistant authentication system that uses an improved version of the Serial Interception Sequence Learning task, employing additional auditory and haptic modalities backed by concepts borrowed from cognitive psychology. We carefully modify the visual stimuli as well as add auditory and haptic stimuli to improve the implicit learning process, resulting in faster training and longer retention. Moreover, our improvements guarantee that explicit recognition of the trained passwords remains suppressed.

Federated learning allows us to distributively train a machine learning model where multiple parties share local model parameters without sharing private data. However, parameter exchange may still leak information. Several approaches have been proposed to overcome this, based on multi-party computation, fully homomorphic encryption, etc.; many of these protocols are slow and impractical for real-world use as they involve a large number of cryptographic operations. In this paper, we propose the use of Trusted Execution Environments (TEE), which provide a platform for isolated execution of code and handling of data, for this purpose. We describe Flatee, an efficient privacy-preserving federated learning framework across TEEs, which considerably reduces training and communication time. Our framework can handle malicious parties (we do not natively solve adversarial data poisoning, though we describe a preliminary approach to handle this).

Affirmative action in the form of reservations is a divisive and contentious topic of policy in India. In this paper, we aim to create a principled and data-driven model to design the reservations policy in India. We look at some arguments against current policy and try to resolve them. We use statistical modeling to create our new framework, RAMSES (Rigorous and Adaptive Measurement of Socio-Economic Status). RAMSES measures the multidimensional disadvantage faced by an individual as an "adjusted income", which attempts to calibrate the quantum of compensatory aid in the form of reservations for that individual to have a level playing field. We illustrate our model using a case study.

Recent developments in secure computation have led to significant improvementsin efficiency and functionality. These efforts created compilers that form thebackbone of practical secure computation research. Unfortunately, many of theartifacts that are being used to demonstrate new research for secure computationare incomplete, incorrect, or unstable, leading to demonstrably erroneousresults and inefficiencies - extending even to the most recently developedcompiler systems. This is a problem because it hampers research and underminesfeasibility tests when other researchers attempt to use these tools. We addressthese problems and present Frigate, a principled compiler and fast circuitinterpreter for secure computation. To ensure correctness we apply bestpractices for compiler design and development, including the use of standarddata structures, helpful negative results, and structured validation testing. Our systematic validation tests include checks on the internal compiler state, combinations of operators, and edge cases based on widely used techniques anderrors we have observed in other work. This produces a compiler that buildscorrect circuits, is efficient and extensible. Frigate creates circuits withgate counts comparable to previous work, but does so with compile time speedupsas high as 447x compared with the best results from previous work in circuit compilers. By creating avalidated tool, our compiler will allow future secure computationimplementations to be developed quickly and correctly.

Two-party secure-function evaluation (SFE) has become significantly more feasible, even on resource-constrained devices, because of advances in server-aided computation systems. However, there are still bottlenecks, particularly in the input-validation stage of a computation. Moreover, SFE research has not yet devoted sufficient attention to the important problem of retaining state after a computation has been performed so that expensive processing does not have to be repeated if a similar computation is done again. This paper presents PartialGC, an SFE system that allows the reuse of encrypted values generated during a garbled-circuit computation. We show that using PartialGC can reduce computation time by as much as 96% and bandwidth by as much as 98% in comparison with previous outsourcing schemes for secure computation. We demonstrate the feasibility of our approach with two sets of experiments, one in which the garbled circuit is evaluated on a mobile device and one in which it is evaluated on a server. We also use PartialGC to build a privacy-preserving ``friend-finder'' application for Android. The reuse of previous inputs to allow stateful evaluation represents a new way of looking at SFE and further reduces computational barriers.

Interdomain routing involves coordination among mutually distrustful parties, leading to the requirements that BGP provide policy autonomy, flexibility, and privacy. BGP provides these properties via the distributed execution of policy-based decisions during the iterative route computation process. This approach has poor convergence properties, makes planning and failover difficult, and is extremely difficult to change. To rectify these and other problems, we propose a radically different approach to interdomain-route computation, based on secure multi-party computation (SMPC). Our approach provides stronger privacy guarantees than BGP and enables the deployment of new policy paradigms. We report on an initial exploration of this idea and outline future directions for research.