arXiv subjects

Aditya Jain

Publications and source records attributed to Aditya Jain.

31 records · Page 2Linked to original sources

arXiv2023

Limits on quantum deletion from no signaling principle

One of the fundamental restrictions that quantum mechanics imposes is the "No deletion Theorem" which tells us that given two identical unknown quantum states, it is impossible to delete one of them. But nevertheless if not perfect, people have tried to delete it approximately. In these approximate deleting processes our basic target is to delete one of the two identical copies as much as possible while preserving the other copy. In this brief report, by using the No communication theorem (NCT) (impossibility of sending signal faster than light using a quantum resource) as a guiding principle, we obtain a bound on the sum of the fidelity of deletion and the fidelity of preservation. Our result not only brings out the complementary relation between these two fidelities but also predicts the optimal value of the fidelity of deletion achievable for a given fidelity of preservation under no signaling constraint. This work eventually saturates the quest for finding out the optimal value of deletion within the NCT framework.

quant-ph↗

arXiv2022

Superconformal invariants and spinning correlators in 3d $\cal{N}=2$ SCFTs

We construct superconformal invariants in superspace which are used to build 3-point correlators of spinning operators in general $\cal{N}=2$ superconformal field theories in three dimensions. Our systematic analysis includes various relations between these invariants and provides a minimal set of parity-even and parity-odd invariants which is further used to construct general 3-point functions in any 3d $\cal{N}=2$ SCFT. For conserved (super)currents, we explicitly compute various 3-point functions using Wick contractions in the free field case, and express them in terms of the constructed parity-even invariants. We give evidence through examples for the claim that the 3-point function of conserved currents generally comprises of two parts - a parity-even piece coming from the free theory, and a parity-odd piece.

hep-th↗

arXiv2022

International cooperation and competition in orbit-use management

Orbit-use management efforts can be structured as binding national regulatory policies or as self-enforcing international treaties. New treaties to control space debris growth appear unlikely in the near future. Spacefaring nations can pursue national regulatory policies, though regulatory competition and open access to orbit make their effectiveness unclear. We develop a game-theoretic model of national regulatory policies and self-enforcing international treaties for orbit-use management in the face of open access, regulatory competition, and catastrophe. While open access limits the effectiveness of national policies, market-access control ensures the policies can improve environmental quality. A large enough stock of legacy debris ensures existence of a global regulatory equilibrium where all nations choose to levy environmental regulations on all satellites. The global regulatory equilibrium supports a self-enforcing treaty to avert catastrophe by making it costlier to leave the treaty and free ride.

econ.TH↗

arXiv2022

muBoost: An Effective Method for Solving Indic Multilingual Text Classification Problem

Text Classification is an integral part of many Natural Language Processing tasks such as sarcasm detection, sentiment analysis and many more such applications. Many e-commerce websites, social-media/entertainment platforms use such models to enhance user-experience to generate traffic and thus, revenue on their platforms. In this paper, we are presenting our solution to Multilingual Abusive Comment Identification Problem on Moj, an Indian video-sharing social networking service, powered by ShareChat. The problem dealt with detecting abusive comments, in 13 regional Indic languages such as Hindi, Telugu, Kannada etc., on the videos on Moj platform. Our solution utilizes the novel muBoost, an ensemble of CatBoost classifier models and Multilingual Representations for Indian Languages (MURIL) model, to produce SOTA performance on Indic text classification tasks. We were able to achieve a mean F1-score of 89.286 on the test data, an improvement over baseline MURIL model with a F1-score of 87.48.

cs.CL↗

arXiv2022

Machine Learning in Sports: A Case Study on Using Explainable Models for Predicting Outcomes of Volleyball Matches

Machine Learning has become an integral part of engineering design and decision making in several domains, including sports. Deep Neural Networks (DNNs) have been the state-of-the-art methods for predicting outcomes of professional sports events. However, apart from getting highly accurate predictions on these sports events outcomes, it is necessary to answer questions such as "Why did the model predict that Team A would win Match X against Team B?" DNNs are inherently black-box in nature. Therefore, it is required to provide high-quality interpretable, and understandable explanations for a model's prediction in sports. This paper explores a two-phased Explainable Artificial Intelligence(XAI) approach to predict outcomes of matches in the Brazilian volleyball League (SuperLiga). In the first phase, we directly use the interpretable rule-based ML models that provide a global understanding of the model's behaviors based on Boolean Rule Column Generation (BRCG; extracts simple AND-OR classification rules) and Logistic Regression (LogReg; allows to estimate the feature importance scores). In the second phase, we construct non-linear models such as Support Vector Machine (SVM) and Deep Neural Network (DNN) to obtain predictive performance on the volleyball matches' outcomes. We construct the "post-hoc" explanations for each data instance using ProtoDash, a method that finds prototypes in the training dataset that are most similar to the test instance, and SHAP, a method that estimates the contribution of each feature on the model's prediction. We evaluate the SHAP explanations using the faithfulness metric. Our results demonstrate the effectiveness of the explanations for the model's predictions.

cs.LG↗

arXiv2021

Solving Fashion Recommendation -- The Farfetch Challenge

Recommendation engines are integral to the modern e-commerce experience, both for the seller and the end user. Accurate recommendations lead to higher revenue and better user experience. In this paper, we are presenting our solution to ECML PKDD Farfetch Fashion Recommendation Challenge. The goal of this challenge is to maximize the chances of a click when the users are presented with set of fashion items. We have approached this problem as a binary classification problem. Our winning solution utilizes Catboost as the classifier and Bayesian Optimization for hyper parameter tuning. Our baseline model achieved MRR of 0.5153 on the validation set. Bayesian optimization of hyper parameters improved the MRR to 0.5240 on the validation set. Our final submission on the test set achieved a MRR of 0.5257.

cs.LG↗

arXiv2021

Efficient diagnostics for quantum error correction

Fault-tolerant quantum computing will require accurate estimates of the resource overhead, but standard metrics such as gate fidelity and diamond distance have been shown to be poor predictors of logical performance. We present a scalable experimental approach based on Pauli error reconstruction to predict the performance of concatenated codes. Numerical evidence demonstrates that our method significantly outperforms predictions based on standard error metrics for various error models, even with limited data. We illustrate how this method assists in the selection of error correction schemes.

quant-ph↗

arXiv2021

Optimizing Cost per Click for Digital Advertising Campaigns

Cost per click is a common metric to judge digital advertising campaign performance. In this paper we discuss an approach that generates a feature targeting recommendation to optimise cost per click. We also discuss a technique to assign bid prices to features without compromising on the number of features recommended. Our approach utilises impression and click stream data sets corresponding to real time auctions that we have won. The data contains information about device type, website, RTB Exchange ID. We leverage data across all campaigns that we have access to while ensuring that recommendations are sensitive to both individual campaign level features and globally well performing features as well. We model Bid recommendation around the hypothesis that a click is a Bernoulli trial and click stream follows Binomial distribution which is then updated based on live performance ensuring week over week improvement. This approach has been live tested over 10 weeks across 5 campaigns. We see Cost per click gains of 16-60% and click through rate improvement of 42-137%. At the same time, the campaign delivery was competitive.

stat.AP↗

arXiv2020

Biased Models Have Biased Explanations

We study fairness in Machine Learning (FairML) through the lens of attribute-based explanations generated for machine learning models. Our hypothesis is: Biased Models have Biased Explanations. To establish that, we first translate existing statistical notions of group fairness and define these notions in terms of explanations given by the model. Then, we propose a novel way of detecting (un)fairness for any black box model. We further look at post-processing techniques for fairness and reason how explanations can be used to make a bias mitigation technique more individually fair. We also introduce a novel post-processing mitigation technique which increases individual fairness in recourse while maintaining group level fairness.

cs.LG↗

arXiv2019

Asymmetric broadcasting of quantum correlations

In this work, we exhaustively investigate $1 \rightarrow 2$ local and nonlocal broadcasting of entanglement as well as correlations beyond entanglement (geometric discord) using asymmetric Pauli cloners with most general two qubit state as the resource. We exemplify asymmetric broadcasting of entanglement using Maximally Entangled Mixed States. We demonstrate the variation of broadcasting range with the amount of entanglement present in the resource state as well as with the asymmetry in the cloner. We show that it is impossible to optimally broadcast geometric discord with the help of these asymmetric Pauli cloning machines. We also study the problem of $1 \rightarrow 3$ broadcasting of entanglement using non-maximally entangled state (NME) as the resource. For this task, we introduce a method we call successive broadcasting which involves application of $1 \rightarrow 2$ optimal cloning machines multiple times. We compare and contrast the performance of this method with the application of direct $1 \rightarrow 3$ optimal cloning machines. We show that $1 \rightarrow 3$ optimal cloner does a better job at broadcasting than the successive application of $1 \rightarrow 2$ cloners and the successive method can be beneficial in the absence of $1 \rightarrow 3$ cloners. We also bring out the fundamental difference between the tasks of cloning and broadcasting in the final part of the manuscript. We create examples to show that there exist local unitaries which can be employed to give a better range for broadcasting. Such unitary operations are not only economical, but also surpass the best possible range obtained using existing cloning machines enabling broadcasting of lesser entangled states. This result opens up a new direction in exploration of methods to facilitate broadcasting which may outperform the standard strategies implemented through cloning transformations.

quant-ph↗

arXiv2016

How local is the information in MPS/PEPS tensor networks?

Two dimensional tensor networks such as projected entangled pairs states (PEPS) are generally hard to contract. This is arguably the main reason why variational tensor network methods in 2D are still not as successful as in 1D. However, this is not necessarily the case if the tensor network represents a gapped ground state of a local Hamiltonian; such states are subject to many constraints and contain much more structure. In this paper we introduce a new approach for approximating the expectation value of a local observable in ground states of local Hamiltonians that are represented as PEPS tensor-networks. Instead of contracting the full tensor-network, we try to estimate the expectation value using only a local patch of the tensor-network around the observable. Surprisingly, we demonstrate that this is often easier to do when the system is frustrated. In such case, the spanning vectors of the local patch are subject to non-trivial constraints that can be utilized via a semi-definite program to calculate rigorous lower- and upper-bounds on the expectation value. We test our approach in 1D systems, where we show how the expectation value can be calculated up to at least 3 or 4 digits of precision, even when the patch radius is smaller than the correlation length.

quant-ph↗

arXiv2014

Large quality factor in sheet metamaterials made from dark dielectric meta-atoms

Metamaterials-or artificial electromagnetic materials-can create media with properties unattainable in nature, but mitigating dissipation is a key challenge for their further development. Here, we demonstrate a low-loss metamaterial by exploiting dark bound states in dielectric inclusions coupled to the external waves by small nonresonant metallic antennas. We experimentally demonstrate a dispersion-engineered metamaterial based on a meta-atom made from alumina, and we show that its resonance has a much larger quality factor than metal-based meta-atoms. Finally, we show that our dielectric meta-atom can be used to create sheet metamaterials with negative permittivity or permeability.

physics.optics↗

arXiv2012

Electromagnetically Induced Transparency and Absorption in Metamaterials: The Radiating Two-Oscillator Model and Its Experimental Confirmation

Several classical analogues of electromagnetically induced transparency in metamaterials have been demonstrated. A simple two-resonator model can describe their absorption spectrum qualitatively, but fails to provide information about the scattering properties-e.g., transmission and group delay. Here we develop an alternative model that rigorously includes the coupling of the radiative resonator to the external electromagnetic fields. This radiating two-oscillator model can describe both the absorption spectrum and the scattering parameters quantitatively. The model also predicts metamaterials with a narrow spectral feature in the absorption larger than the background absorption of the radiative element. This classical analogue of electromagnetically induced absorption is shown to occur when both the dissipative loss of the radiative resonator and the coupling strength are small. These predictions are subsequently demonstrated in experiments.

physics.optics↗