Filters








419 Hits in 2.0 sec

Lexical State Analyzer [article]

Kartik Gupta, V. Krishna Nandivada
2013 arXiv   pre-print
Lexical states provide a powerful mechanism to scan regular expressions in a context sensitive manner. At the same time, lexical states also make it hard to reason about the correctness of the grammar. We first categorize the related correctness issues into two classes: errors and warnings, and then present a context sensitive and a context insensitive analysis to identify errors and warnings in context-free-grammars (CFGs). We also present a comparative study of these analyses. A standalone
more » ... l (LSA) has also been implemented by us that can identify errors and warnings in JavaCC grammars. The LSA tool outputs a graph that depicts the grammar and the error transitions. It can also generates counter example strings that can be used to establish the errors. We have used LSA to analyze a host of open-source JavaCC grammar files to good effect.
arXiv:1308.3156v1 fatcat:w2kqigu2xnbw3am5rqe4wonw5y

Learning Minimax Estimators via Online Learning [article]

Kartik Gupta, Arun Sai Suggala, Adarsh Prasad, Praneeth Netrapalli, Pradeep Ravikumar
2020 arXiv   pre-print
We consider the problem of designing minimax estimators for estimating the parameters of a probability distribution. Unlike classical approaches such as the MLE and minimum distance estimators, we consider an algorithmic approach for constructing such estimators. We view the problem of designing minimax estimators as finding a mixed strategy Nash equilibrium of a zero-sum game. By leveraging recent results in online learning with non-convex losses, we provide a general algorithm for finding a
more » ... xed-strategy Nash equilibrium of general non-convex non-concave zero-sum games. Our algorithm requires access to two subroutines: (a) one which outputs a Bayes estimator corresponding to a given prior probability distribution, and (b) one which computes the worst-case risk of any given estimator. Given access to these two subroutines, we show that our algorithm outputs both a minimax estimator and a least favorable prior. To demonstrate the power of this approach, we use it to construct provably minimax estimators for classical problems such as estimation in the finite Gaussian sequence model, and linear regression.
arXiv:2006.11430v1 fatcat:ieqgr3bfcrcw7nm4c5z6mthuuu

Problems with automating translation of movie/TV show subtitles [article]

Prabhakar Gupta, Mayank Sharma, Kartik Pitale, Keshav Kumar
2019 arXiv   pre-print
We present 27 problems encountered in automating the translation of movie/TV show subtitles. We categorize each problem in one of the three categories viz. problems directly related to textual translation, problems related to subtitle creation guidelines, and problems due to adaptability of machine translation (MT) engines. We also present the findings of a translation quality evaluation experiment where we share the frequency of 16 key problems. We show that the systems working at the
more » ... of Natural Language Processing do not perform well for subtitles and require some post-processing solutions for redressal of these problems
arXiv:1909.05362v1 fatcat:4cknpukntfgfzmttmcnwbtzkji

Plexiform Neurofibroma of Nasal Tip

Rajanala Venkata Nataraj, Mohan Jagade, Kartik Parelkar, Reshma Hanawte, Arpita Singhal, Dev Rengaraja, Kiran Kulsange, Kartik Rao, Pallavi Gupta
2015 International Journal of Otolaryngology and Head & Neck Surgery  
Introduction The first reported cases of external nasal neurogenic tumors were reported by New and Devine in 1947 and Das Gupta et al. in 1969 [1] [2] .  ... 
doi:10.4236/ijohns.2015.46068 fatcat:djdf4dtfqbhrdoa3l7gyu3a234

True leiomyoma of prostate

Kartik Chandrakant Gupta
2021 The Sri Lanka Journal of Surgery  
Correspondence: Kartik Chandrakant Gupta E-mail: drkartik.gupta@yahoo.in https://orcid.org/0000-0002-3462-0959 Received: 29-11-2020 Accepted: 19-03-2021 DOI: http://doi.org/10.4038/sljs.v39i1.8744  ... 
doi:10.4038/sljs.v39i1.8744 fatcat:qnk3wd24wraybmf37smxowewvq

Augmentation Grafts in Septorhinoplasty: Our Experience

R. V. Nataraj, Jagade Mohan, Chavan Reshma, Parelkar Kartik, Hanawte Reshma, Singhal Arpita, Kulsange Kiran, Rengaraja Dev, Rao Kartik, Gupta Pallavi
2015 International Journal of Otolaryngology and Head & Neck Surgery  
Augmentation of nasal tip and/or dorsum forms the keystone of any Septorhinoplasty surgery. The grafts available for augmentation are numerous and varied. Choice of the graft depends upon the type of augmentation required, patient characteristics and, most importantly, the surgeon. In this article, we would like to present our experience with various augmentation grafts. In our experience, autografts are best grafts for augmentation. But in cases of revision surgeries or deficiency of
more » ... , allografts can be used. Our choice of allograft is Poly Diaxone Sheath or PDS.
doi:10.4236/ijohns.2015.44054 fatcat:3lvrcvxqhbd7zl366v2zbn7edi

Classifying Object Manipulation Actions based on Grasp-types and Motion-Constraints [article]

Kartik Gupta, Darius Burschka, Arnav Bhavsar
2018 arXiv   pre-print
In this work, we address a challenging problem of fine-grained and coarse-grained recognition of object manipulation actions. Due to the variations in geometrical and motion constraints, there are different manipulations actions possible to perform different sets of actions with an object. Also, there are subtle movements involved to complete most of object manipulation actions. This makes the task of object manipulation action recognition difficult with only just the motion information. We
more » ... ose to use grasp and motion-constraints information to recognise and understand action intention with different objects. We also provide an extensive experimental evaluation on the recent Yale Human Grasping dataset consisting of large set of 455 manipulation actions. The evaluation involves a) Different contemporary multi-class classifiers, and binary classifiers with one-vs-one multi- class voting scheme, b) Differential comparisons results based on subsets of attributes involving information of grasp and motion-constraints, c) Fine-grained and Coarse-grained object manipulation action recognition based on fine-grained as well as coarse-grained grasp type information, and d) Comparison between Instance level and Sequence level modeling of object manipulation actions. Our results justifies the efficacy of grasp attributes for the task of fine-grained and coarse-grained object manipulation action recognition.
arXiv:1806.07574v1 fatcat:4vtq63rgxbbkjlr5zbalw2oufe

CullNet: Calibrated and Pose Aware Confidence Scores for Object Pose Estimation [article]

Kartik Gupta, Lars Petersson, Richard Hartley
2019 arXiv   pre-print
We present a new approach for a single view, image-based object pose estimation. Specifically, the problem of culling false positives among several pose proposal estimates is addressed in this paper. Our proposed approach targets the problem of inaccurate confidence values predicted by CNNs which is used by many current methods to choose a final object pose prediction. We present a network called CullNet, solving this task. CullNet takes pairs of pose masks rendered from a 3D model and cropped
more » ... egions in the original image as input. This is then used to calibrate the confidence scores of the pose proposals. This new set of confidence scores is found to be significantly more reliable for accurate object pose estimation as shown by our results. Our experimental results on multiple challenging datasets (LINEMOD and Occlusion LINEMOD) reflects the utility of our proposed method. Our overall pose estimation pipeline outperforms state-of-the-art object pose estimation methods on these standard object pose estimation datasets. Our code is publicly available on https://github.com/kartikgupta-at-anu/CullNet.
arXiv:1909.13476v1 fatcat:mhhtvg275vc5bdh2syxhd2k7uq

Biomarkers and Outcomes in Hospitalized Covid-19 Patients: A Prospective Registry [article]

Raghubir S Khedar, Rajeev Gupta, Krishna Kumar Sharma, Kartik Mittal, Harshad C Ambaliya, Jugal B Gupta, Surendra Singh, Swati Sharma, Yogendra Singh, Alok Mathur
2022 medRxiv   pre-print
Objectives: To determine association of biomarkers high sensitivity C-reactive protein (hsCRP), D-dimer, interleukin-6 (IL-6), lactic dehydrogenase (LDH), ferritin and neutrophil-lymphocyte ratio (NLR) at hospital admission with clinical features and outcomes in Covid-19. Methods: Successive virologically confirmed Covid-19 patients hospitalized from April 2020 to July 2021 were recruited in a prospective registry. Details of clinical presentation, investigations, management and outcomes were
more » ... corded. All the biomarkers were divided into tertiles to determine associations with clinical features and outcomes. Numerical data are presented in median and interquartile range (IQR 25-75). Univariate and multivariate (age, sex, risk factor, comorbidity adjusted) odds ratio (OR) and 95% confidence intervals (CI) were calculated to determine association of deaths with each biomarker. Results: We identified 3036 virologically confirmed Covid-19 patients during the study period, 1215 were hospitalized and included in the present study. Men were 70.0%, aged >60y 44.8%, hypertension 44.8% diabetes 39.6% and cardiovascular disease 18.9%. Median symptom duration was 5 days (IQR 4-7) and SpO2 95% (90-97). Total white cell count was 6.9x103/micro-litre, (5.0-9.8), neutrophils 79.2% (68.1-88.2) and lymphocytes 15.8% (8.7-25.5). Medians (IQR) for biomarkers were hsCRP 6.9 mg/dl (2.2-18.9), D-dimer 464 ng/dl (201-982), IL-6 20.1 ng/dl (6.5-60.4), LDH 284 mg/dl (220-396) and ferritin 351 mg/dl (159-676). Oxygen support at admission was in 38.6%, and non-invasive or invasive ventilatory support in 11.0% and 11.6% respectively. 173 (13.9%) patients died and 15 (1.2%) transferred to hospice care. For each biomarker, those in the second and third tertiles, compared to the first, had worse clinical and laboratory abnormalities, and greater oxygen and ventilatory support. Multivariate adjusted OR (95% CI) for deaths in second and third vs first tertiles, respectively, were for hsCRP 2.29(1.14-4.60) and 13.39(7.23-24.80); D-dimer 3.26(1.31-7.05) and 13.89(6.87-28.27); IL-6 2.61(1.31-5.18) and 10.96(5.88-20.43); ferritin 3.19(1.66-6.11) and 9.13(4.97-16.78); LDH 1.85(0.87-3.97) and 10.51(5.41-20.41); and NLR 3.34(1.62-6.89) and 17.52(9.03-34.00) (p<0.001). Conclusions: In Covid-19, high levels of biomarkers- hsCRP, D-dimer, IL-6, LDH, ferritin and NLR are associated with more severe illness and significantly greater in-hospital mortality. NLR, a simple, widely available and inexpensive investigation provides prognostic information similar to the more expensive biomarkers.
doi:10.1101/2022.07.20.22277718 fatcat:asncvmwn45fqdlkkp2be74fgom

Nearly-optimal Robust Matrix Completion [article]

Yeshwanth Cherapanamjeri, Kartik Gupta, Prateek Jain
2016 arXiv   pre-print
In this paper, we consider the problem of Robust Matrix Completion (RMC) where the goal is to recover a low-rank matrix by observing a small number of its entries out of which a few can be arbitrarily corrupted. We propose a simple projected gradient descent method to estimate the low-rank matrix that alternately performs a projected gradient descent step and cleans up a few of the corrupted entries using hard-thresholding. Our algorithm solves RMC using nearly optimal number of observations as
more » ... well as nearly optimal number of corruptions. Our result also implies significant improvement over the existing time complexity bounds for the low-rank matrix completion problem. Finally, an application of our result to the robust PCA problem (low-rank+sparse matrix separation) leads to nearly linear time (in matrix dimensions) algorithm for the same; existing state-of-the-art methods require quadratic time. Our empirical results corroborate our theoretical results and show that even for moderate sized problems, our method for robust PCA is an an order of magnitude faster than the existing methods.
arXiv:1606.07315v3 fatcat:syxxscwky5f5lhes2dkmgu44ci

Improved Gradient based Adversarial Attacks for Quantized Networks [article]

Kartik Gupta, Thalaiyasingam Ajanthan
2021 arXiv   pre-print
Neural network quantization has become increasingly popular due to efficient memory consumption and faster computation resulting from bitwise operations on the quantized networks. Even though they exhibit excellent generalization capabilities, their robustness properties are not well-understood. In this work, we systematically study the robustness of quantized networks against gradient based adversarial attacks and demonstrate that these quantized models suffer from gradient vanishing issues
more » ... show a fake sense of robustness. By attributing gradient vanishing to poor forward-backward signal propagation in the trained network, we introduce a simple temperature scaling approach to mitigate this issue while preserving the decision boundary. Despite being a simple modification to existing gradient based adversarial attacks, experiments on multiple image classification datasets with multiple network architectures demonstrate that our temperature scaled attacks obtain near-perfect success rate on quantized networks while outperforming original attacks on adversarially trained models as well as floating-point networks. Code is available at https://github.com/kartikgupta-at-anu/attack-bnn.
arXiv:2003.13511v2 fatcat:7m5afvgqujbfnhiifzvx6vb57q

Quantization Robust Federated Learning for Efficient Inference on Heterogeneous Devices [article]

Kartik Gupta, Marios Fournarakis, Matthias Reisser, Christos Louizos, Markus Nagel
2022 arXiv   pre-print
Federated Learning (FL) is a machine learning paradigm to distributively learn machine learning models from decentralized data that remains on-device. Despite the success of standard Federated optimization methods, such as Federated Averaging (FedAvg) in FL, the energy demands and hardware induced constraints for on-device learning have not been considered sufficiently in the literature. Specifically, an essential demand for on-device learning is to enable trained models to be quantized to
more » ... us bit-widths based on the energy needs and heterogeneous hardware designs across the federation. In this work, we introduce multiple variants of federated averaging algorithm that train neural networks robust to quantization. Such networks can be quantized to various bit-widths with only limited reduction in full precision model accuracy. We perform extensive experiments on standard FL benchmarks to evaluate our proposed FedAvg variants for quantization robustness and provide a convergence analysis for our Quantization-Aware variants in FL. Our results demonstrate that integrating quantization robustness results in FL models that are significantly more robust to different bit-widths during quantized on-device inference.
arXiv:2206.10844v1 fatcat:qarpnjiaj5ed5pcknfclgt6cve

Comparative Study of Various Frequent Pattern Mining Algorithms

Amit Mittal, Ashutosh Nagar, Kartik Gupta, Rishi Nahar
2015 IJARCCE  
Frequent pattern mining has become an important data mining task and has been a focused theme in data mining research. Frequent pattern mining aims to find frequently occurring subsets in sequence of sets. The frequent pattern mining appears as a sub problem in many other data mining fields such as association rules discovery, classification, clustering, web mining, market analysis etc. Different frameworks have been defined for frequent pattern mining. The most common one is the support based
more » ... ramework, in which item sets with frequency above a given threshold are found. This paper presents review of different frequent mining algorithms including Apriori, FPgrowth and DIC. A brief description of each technique has been provided. In the last, different frequent pattern mining techniques are compared based on various parameters of importance.
doi:10.17148/ijarcce.2015.44127 fatcat:pki32giv65aeznsbclkkszr7rm

Plantosphere: Next Generation Adaptive and Smart Agriculture System

Abhishek Verma, Manas Agrawal, Kartik Gupta, Aatif Jamshed, Anurag Mishra, Harsh Khatter, Gopal Gupta, Sanjeev Chandra Neupane, Paulo Jorge Sequeira Gonçalves
2022 Journal of Sensors  
Around 75% of the population in India is engaged in agriculture and farming. The sustainability of every economy is based on agriculture. It has a major influence on financial growth and fundamental transformation in the long run. Artificial intelligence will usher in a revolution in agricultural operations in the future. This revolution has protected crops from being negatively affected by a variety of factors such as climate change, soil porosity, and water availability. Crop monitoring, soil
more » ... management, and insect identification, to name a few examples, are all conceivable uses of artificial intelligence in agriculture. The primary purpose of artificial intelligence is to close the knowledge gap that exists between inventors and farmers. Detecting disease and monitoring plant health are two of the most difficult challenges in sustainable farming. As a result, image processing technology must be used to detect plant sickness at an early stage. Photographic capture, preprocessing, segmentation, feature extraction, and sickness categorization are all part of the procedure. In reality, computer image processing was used long before human eyes were able to detect the signs and symptoms of the disease. Taking into account the climatic conditions in various parts of the world. Climate change directly affects crop output. Several soil and atmospheric characteristics are detected to anticipate the optimal crop. Sedimentation is measured by soil parameters such as pH and moisture. Today, a platform that allows farmers to advertise their products is in high demand. This paper proposes a system where farmers sell directly to clients, bypassing wholesalers and traders. A predictive analytics solution is required to maximize the farmer's profit.
doi:10.1155/2022/5421312 fatcat:rtflma5evbf65ecwaqmcmrsxpa

FL Games: A federated learning framework for distribution shifts [article]

Sharut Gupta and Kartik Ahuja and Mohammad Havaei and Niladri Chatterjee and Yoshua Bengio
2022 arXiv   pre-print
Federated learning aims to train predictive models for data that is distributed across clients, under the orchestration of a server. However, participating clients typically each hold data from a different distribution, whereby predictive models with strong in-distribution generalization can fail catastrophically on unseen domains. In this work, we argue that in order to generalize better across non-i.i.d. clients, it is imperative to only learn correlations that are stable and invariant across
more » ... domains. We propose FL Games, a game-theoretic framework for federated learning for learning causal features that are invariant across clients. While training to achieve the Nash equilibrium, the traditional best response strategy suffers from high-frequency oscillations. We demonstrate that FL Games effectively resolves this challenge and exhibits smooth performance curves. Further, FL Games scales well in the number of clients, requires significantly fewer communication rounds, and is agnostic to device heterogeneity. Through empirical evaluation, we demonstrate that FL Games achieves high out-of-distribution performance on various benchmarks.
arXiv:2205.11101v1 fatcat:zpnp7mjsozaorecamjldunx6ii
« Previous Showing results 1 — 15 out of 419 results