Abstract

Unmanned Aerial Vehicle (UAV)-enabled mobile edge computing (MEC) is considered to offer computational capabilities to the resource-constraints end-users. In this paper, we study the task offloading strategy in UAV-enabled MEC systems, where end-users offload the computation-intensive tasks to the UAV to minimize the overall cost in terms of the weighted delay and energy consumption. The end-users either process the task by itself or offload the tasks to the UAV that acts as a computing access point. However, due to the computation bottleneck and limited channel capacity between UAV and the end-users, it becomes a challenging issue to offload the entire tasks to the UAV. Thus, to find the optimal offloading decision for the tasks generated by the end-users, we build a distributed deep neural network (DNN). In the proposed distributed DNN model, we train multiple DNNs in the same training instance, and finally, for validation, we select the DNN that gives the least training loss. For faster convergence of the training process, we use the optimal generated offloading decision using a Quadratically Constrained Linear Program (QCLP) with Semidefinite Relaxation (SDR). The extensive simulation results show that the offloading decision produced by the trained DNN can achieve near-optimal performance with numerous system parameter settings.

Abstract

With the rising need for reliable and real-time pose estimation in resource constrained environments such as smartphones, IoT devices, and head mounts, we need an efficient and compact pose estimation framework. To this end, we propose PoseFromGraph1, a light-weight 3D pose estimation framework. The inputs to PoseFromGraph are: a graph obtained by skeletonizing the 3D meshes using the prairie-fire analogy and the RGB image, and the output is the 3D pose of the object. The introduction of 3D shapes to the architecture makes our model category-agnostic. Unlike computationally expensive multi-view geometry and point-cloud based representations to estimate pose, our approach uses a message passing network to incorporate local neighborhood information at the same time maintaining global shape property in a graph by optimizing a neighborhood preserving objective. PoseFromGraph surpasses the state-of-the-art pose estimation methods in terms of accuracy achieving, 84.43% on the Pascal3D dataset, and at the same time yields 4 × reduction in the space and time complexity. The compact pose estimation models can then be used to facilitate on-device inference in applications in Augmented Reality and Robotics for 3D virtual model overlay.

Abstract

3While deep learning approaches have shown promising results in Natural Language Processing and Computer Vision domains, they have not yet been able to achieve impressive results in Ontology Alignment, and have typically performed worse than rule-based approaches. Some of the major reasons for this are: a) poor modelling of context, b) overfitting of standard DL models, and c) dataset sparsity, caused by class imbalance of positive alignment pairs wrt negative pairs. To mitigate these limitations, we propose a dual-attention based approach that uses a multi-faceted context representation to compute contextualized representations of concepts, which is then used to discover semantically equivalent concepts.

Abstract

Few-shot segmentation (FSS) methods perform image segmentation for a particular object class in a target (query) image, using a small set of (support) image-mask pairs. Recent deep neural network based FSS methods leverage high-dimensional feature similarity between the foreground features of the support images and the query image features. In this work, we demonstrate gaps in the utilization of this similarity information in existing methods, and present a framework - SimPropNet, to bridge those gaps. We propose to jointly predict the support and query masks to force the support features to share characteristics with the query features. We also propose to utilize similarities in the background regions of the query and support images using a novel foreground-background attentive fusion mechanism. Our method achieves stateof-the-art results for one-shot and five-shot segmentation on the PASCAL-5 i dataset. The paper includes detailed analysis and ablation studies for the proposed improvements and quantitative comparisons with contemporary methods.

Abstract

This paper describes the participation of team F1toF6 (LTRC, IIIT-Hyderabad) for the WMT 2020 task, similar language translation. We experimented with attention based recurrent neural network architecture (seq2seq) for this task. We explored the use of different linguistic features like POS and Morph along with back translation for Hindi-Marathi and Marathi-Hindi machine translation.

Abstract

The adoption of Machine Learning (ML) in software applications has increased in domains like healthcare, banking and others. leading to coining of the term MLWare applications. However, challenges like diverse code base, complex components, lack of expertise, etc. persist in development and maintenance of these applications. Application of software engineering patterns and practices for the development of MLWare applications can improve maintainability, extensibility, scalability and other software quality parameters. In this paper, we propose an approach for developing MLWare applications using a pattern oriented approach. We demonstrate the approach on a credit risk scorecard application that can helps loan officer identify risk patterns and make loan decisions.

Abstract

Software quality has become the lever of differentiation in today’s competitive marketplace. Quality at speed is the customer demand and automation is the biggest bottleneck holding the evolution of quality function. Increased levels of automation and intelligence in software engineering are the emerging trends across the IT field. As systems and software processes guide the life cycle activities and are the vehicles for building quality, it is necessary to look at the process infrastructure for the extent of process automation support provided and the digital enablement. This paper maps out the existing process infrastructure support in industry practice and proposes a roadmap for digital re-imagination of software and systems processes. Harmonizing the quality engineering themes with digital technologies, we propose a framework for building an intelligent software process infrastructure, iSPIN that can help in digital re-imagination of software and systems lifecycle processes. The framework has been implemented using digital technologies and has been piloted with one of the industry business unit for re- imagination of “proposal process”. The proposed iSPIN framework will help in unprecedented automation and quality engineering at each process step and paves the way towards realizing the dictums of “Quality at Speed” and “Digital transformation of Software Process”.

Abstract

Negotiation is an important component of the interaction process among humans. With increasing automation, autonomous agents are expected to take over a lot of this interaction process. Much of automated negotiation literature focuses on agents having a static and known reservation value. In situations involving dynamic environments e.g., an agent negotiating on behalf of a human regarding a meeting, agents can have a reservation value (RV) that is a function of time. This leads to a different set of challenges that may need additional reasoning about the concession behavior. In this paper, we build upon Negotiation algorithms such as ONAC (Optimal Non-Adaptive Concession) and TimeDependent Techniques such as Boulware which work on settings where the reservation value of the agent is fixed and known. Although these algorithms can encode dynamic RV, their concession behavior and hence the properties they were expected to display would be different from when the RV is static, even though the underlying negotiation algorithm remains the same. We, therefore, propose to use one of Counter, Bayesian Learning with Regression Analysis or LSTM model on top of each algorithm to develop the PredictRV strategy and show that PredictRV indeed performs better on two different metrics tested on two different domains on a variety of parameter settings. Keywords: Automated negotiation · Dynamic reservati

Abstract

Memes are steadily taking over the feeds of the public on social media. There is always the threat of malicious users on the internet posting offensive content, even through memes. Hence, the automatic detection of offensive images/memes is imperative along with detection of offensive text. However, this is a much more complex task as it involves both visual cues as well as language understanding and cultural/context knowledge. This paper describes our approach to the task of SemEval-2020 Task 8: Memotion Analysis. We chose to participate only in Task A which dealt with Sentiment Classification, which we formulated as a text classification problem. Through our experiments, we explored multiple training models to evaluate the performance of simple text classification algorithms on the raw text obtained after running OCR on meme images. Our submitted model achieved an accuracy of 72.69% and exceeded the existing baseline’s Macro F1 score by 8% on the official test dataset. Apart from describing our official submission, we shall elucidate how different classification models respond to this task.

Abstract

OCR for printed classical Indic documents written in Sanskrit is a challenging research problem. It involves complexities such as image degradation, lack of datasets and long-length words. Due to these challenges, the word accuracy of available OCR systems, both academic and industrial, is not very high for such documents. To address these shortcomings, we develop a Sanskrit specific OCR system. We present an attention-based LSTM model for reading Sanskrit characters in line images. We introduce a dataset of Sanskrit document images annotated at line level. To augment real data and enable high performance for our OCR, we also generate synthetic data via curated font selection and rendering designed to incorporate crucial glyph substitution rules. Consequently, our OCR achieves a word error rate of 15.97% and a character error rate of 3.71% on challenging Indic document texts and outperforms strong baselines. Overall, our contributions set the stage for application of OCRs on large corpora of classic Sanskrit texts containing arbitrarily long and highly conjoined words.