Abstract

This study aims to understand the relationship between personality traits and the emotions, as represented in the Valence-Arousal (VA) space, conveyed through lyrics of their preferred music.

Abstract

This study aims to first formulate a tool to identify an individual’s natural affinity towards appreciation of music vs lyrics and identify personality traits (including empathic and emotional traits) associated with the same.

Abstract

The localization of acoustic sources is a parameter estimation problem where the parameters of interest are the direction of arrivals (DOAs). The DOA estimation problem can be formulated as a sparse parameter estimation problem and solved using compressive sensing (CS) methods. In this paper, the CS method of sparse Bayesian learning (SBL) is used to find the DOAs. We specifically use multi-frequency SBL leading to a non-convex optimization problem, which is solved using fixed-point iterations. We evaluate SBL along with traditional DOA estimation methods of conventional beamforming (CBF) and multiple signal classification (MUSIC) on various source localization tasks from the open access LOCATA dataset. The comparative study shows that SBL significantly outperforms CBF and MUSIC on all the considered tasks.

Abstract

We demonstrate time-, space-, and weight-based calibration techniques for pressure monitoring in the low-cost plastic optical fiber (POF) sensor carpet. It is found that the POFbased pressure sensing platform has several limitations, such as output voltage variations and difference in the sensor response for horizontal and vertical fibers due to convex and concave shapes of bend. To surpass the these limitations, we have developed time-based calibration using time windowing technique to reduce the standard deviation of photodiode output voltage variations by 98%. We have also described space and weight calibration techniques based on the convex and concave shapes of fiber bend. These techniques help in improving the Landweber image reconstruction algorithm to obtain significant clarity and improvement in object pressure images. We also report an artificial intelligence-based algorithm to determine the accuracy of positioning of the load. Experimental results demonstrate that this algorithm gives a mean square error of 0.875 cm in position detection on the carpet. We discuss the potential for a compact and cost-effective pressure sensor carpet, which integrates with the living environment and the outside world. © 2021 Society of Photo-Optical Instrumentation Engineers (SPIE) [DOI:

Abstract

In this paper, we propose deep architectures for learning instance specific abstain (reject option) binary classifiers. The proposed approach uses double sigmoid loss function as described by Kulin Shah and Naresh Manwani in (" Online Active Learning of Reject Option Classifiers", AAAI, 2020), as a performance measure. We show that the double sigmoid loss is classification calibrated. We also show that the excess risk of 0-d-1 loss is upper bounded by the excess risk of double sigmoid loss. We derive the generalization error bounds for the proposed architecture for reject option classifiers. To show the effectiveness of the proposed approach, we experiment with several real world datasets. We observe that the proposed approach not only performs comparable to the state-of-the-art approaches, it is also robust against label noise. We also provide visualizations to observe the important features learned by the network corresponding to the abstaining decision.

Abstract

Civic crowdfunding (CC) is a popular medium for raising funds for civic projects from interested agents. With Blockchains gaining traction, we can implement CC in a reliable, transparent, and secure manner with smart contracts (SCs). The fundamental challenge in CC is free-riding. PPR, the proposal by Zubrickas [21] of giving refund bonus to the contributors, in the case of the project not getting provisioned, has attractive properties. However, as observed by Chandra et al. [6], PPR faces a challenge wherein the agents defer their contribution until the deadline. We define this delaying of contributions as a race condition. To address this, their proposal, PPS, considers the temporal aspects of a contribution. However, PPS is computationally complex, expensive to implement as an SC, and it being sophisticated, it is difficult to explain to a layperson. In this work, our goal is to identify all essential properties a refund bonus scheme must satisfy in order to curb free-riding while avoiding the race condition. We prove Contribution Monotonicity and Time Monotonicity are sufficient conditions for this. We propose three elegant refund bonus schemes satisfying these two conditions leading to three novel mechanisms for CC - PPRG, PPRE, and PPRP. We show that PPRG is the most cost-effective mechanism when deployed as an SC. We show that under certain modest assumptions on valuations of the agents, in PPRG, the project is funded at equilibrium.

Abstract

We consider the problem of achieving fair classification in Federated Learning (FL) under data heterogeneity. Most of the approaches proposed for fair classification require diverse data that represent the different demographic groups involved. In contrast, it is common for each client to own data that represents only a single demographic group. Hence the existing approaches cannot be adopted for fair classification models at the client level. To resolve this challenge, we propose several aggregation techniques. We empirically validate these techniques by comparing the resulting fairness metrics and accuracy on CelebA, UTK, and FairFace datasets.

Abstract

In resource allocation, it is common to assume that agents have a utility for their allocated items and zero utility for unallocated ones. We refer to such valuation domain as 1-D. This assumption of zero utility for unallocated items is not always valid. For example, in the pandemic, allocation of ventilators, oxygen beds, and critical medical help yields dis-utility to an agent when not received in time, i.e., a setting where people consume resources at the cost of others’ utility. Various externalities affect an agent’s utility, i.e., when an agent doesn’t receive an item, it can result in their gain (positive externalities) or loss (negative externalities). The existing preference models lack capturing the setting with these externalities. We conduct a study on a 2-D domain, where each agent has a utility (푣) for an item assigned to it and utility (푣 ′ ) for an item not allocated to it. We consider a generalized model, i.e., goods and chores. There is vast literature for allocating resources both fairly and efficiently. We observe that adapting the existing notions of fairness and efficiency to the 2-D domain is non-trivial. We propose a utility transformation (푇푢) and valuation transformation (푇푣) to convert from the 2-D domain to 1-D, i.e., the existing domain. We study the retention of fairness and efficiency property given this transformation, i.e., an allocation with property P in a 1-D domain also satisfies property P in 2-D, and vice versa. If a property is retainable, we can apply the transformation, and all the existing approaches are valid for the 2-D domain. Further, we study whether we can apply current results in a 2-D domain when they do not retain. We explore fairness notions such as Envy-freeness (EF), Equitability (EQ), Maxmin Share (MMS), and Proportionality and efficiency notions such as Pareto Optimality, Utilitarian Welfare, Nash Welfare, and Egalitarian Welfare.

Abstract

Deep learning’s unprecedented success raises several ethical concerns ranging from biased predictions to data privacy. Researchers tackle these issues by introducing fairness metrics, or federated learning, or differential privacy. A first, this work presents an ethical federated learning model, incorporating all three measures simultaneously. Experiments on the Adult, Bank and Dutch datasets highlight the resulting “empirical interplay” between accuracy, fairness, and privacy.

Abstract

In this paper, we introduce ballooning multi-armed bandits (BL-MAB), a novel extension of the classical stochastic MAB model. In the BL-MAB model, the set of available arms grows (or balloons) over time. In contrast to the classical MAB setting where the regret is computed with respect to the best arm overall, the regret in a BL-MAB setting is computed with respect to the best available arm at each time. We first observe that the existing stochastic MAB algorithms result in linear regret for the BL-MAB model. We prove that, if the best arm is equally likely to arrive at any time instant, a sub-linear regret cannot be achieved. Next, we show that if the best arm is more likely to arrive in the early rounds, one can achieve sub-linear regret. Our proposed algorithm determines (1) the fraction of the time horizon for which the newly arriving arms should be explored and (2) the sequence of arm pulls in the exploitation phase from among the explored arms. Making reasonable assumptions on the arrival distribution of the best arm in terms of the thinness of the distribution’s tail, we prove that the proposed algorithm achieves sublinear instance-independent regret. We further quantify explicit dependence of regret on the arrival distribution parameters. We reinforce our theoretical findings with extensive simulation results. We conclude by showing that our algorithm would achieve sub-linear regret even if (a) the distributional parameters are not exactly known, but are obtained using a reasonable learning mechanism or (b) the best arm is not more likely to arrive early, but a large fraction of arms is likely to arrive relatively early.