Abstract

The pairing of natural language sentences with knowledge graph triples is essential for many downstream tasks like data-to-text generation, facts extraction from sentences (semantic parsing), knowledge graph completion, etc. Most existing methods solve these downstream tasks using neural-based end-to-end approaches that require a large amount of well-aligned training data, which is difficult and expensive to acquire. Recently various unsupervised techniques have been proposed to alleviate this alignment step by automatically pairing the structured data (knowledge graph triples) with textual data. However, these approaches are not well suited for low resource languages that provide two major challenges: (1) unavailability of pair of triples and native text with the same content distribution and (2) limited Natural language Processing (NLP) resources. In this paper, we address the unsupervised pairing of knowledge graph triples with sentences for low resource languages, selecting Hindi as the low resource language. We propose cross-lingual pairing of English triples with Hindi sentences to mitigate the unavailability of content overlap. We propose two novel approaches: NER-based filtering with Semantic Similarity and Key-phrase Extraction with Relevance Ranking. We use our best method to create a collection of 29224 well-aligned English triples and Hindi sentence pairs. Additionally, we have also curated 350 human-annotated golden test datasets for evaluation. We make the code and dataset publicly available.

Abstract

Shape Memory Alloy (SMA) wire actuator contracts or expands in response to heat due to its mechanical (NiTi) property. This paper is about the control perspective of light weight artificial muscle arm based on helical SMA actuator. The Arm is structured around the idea of simplicity and modularity. The post decomposition of the arm dynamics into position and force subspaces, Sliding Mode Control is utilized with the end goal that time taken to reach the equilibrium point from any initial state is guaranteed to be finite time. This setup presents a light weight (70 g) and noiseless operation with a maximum payload of 64 g which is noteworthy when compared to the weight of SMA based artificial muscle and the arm.

Abstract

This paper describes our team oneNLP’s (LTRC, IIIT-Hyderabad) participation for the SEPP-NLG 2021 tasks1, Sentence End and Punctuation Prediction in NLG Text-2021. We applied sequence to tag prediction over contextual embedding as fine-tuning for both of these tasks. We also explored the use of multilingual Bert and multitask learning for these tasks on English, German, French and Italian.

Abstract

In this paper, we (team - oneNLP-IIITH) describe our Neural Machine Translation approaches for English-Marathi (both direction) for LoResMT-20211 . We experimented with transformer based Neural Machine Translation and explored the use of different linguistic features like POS and Morph on subword unit for both English-Marathi and Marathi-English. In addition, we have also explored forward and backward translation using web-crawled monolingual data. We obtained 22.2 (overall 2 nd) and 31.3 (overall 1 st) BLEU scores for English-Marathi and Marathi-English on respectively

Abstract

Introducing the fundamentals of software systems to early undergraduates is a tedious journey for the instructors. It is crucial for students as it lays the foundation and establishes a perception of the upcoming computing courses. With rapid advancements in technology and considering the changes in workforce practices, it is required to upgrade the course curriculum to up-skill the students in contrast to prevailing times. This paper discusses our journey towards upgrading a decade-old introductory course on Software Systems for early undergraduate students. We present our approach towards updating the course curriculum and its delivery modes to ease learnability among early undergraduate students. We implemented mixed pedagogy methods over the years and captured student feedback to evaluate our approach.

Abstract

The avenue of deep learning is scarcely explored in the domain of thermal imaging. Recovering a high-resolution output from images and videos is a classical problem in many computer vision applications. In this paper, we propose an optimized pipeline for a real-time video super-resolution task using thermal camera on embedded edge device. To tackle the challenges, we make contributions in the following several aspects: 1) comparative study of selected deep learning super-resolution models; 2) constructing and optimizing an end-to-end inference pipeline; 3) using cutting edge technology to integrate the whole workflow; 4) a real-time performance was achieved using less data; 5) we have also experimented the entire pipeline on our custom thermal dataset. As a consequence, the chosen model was able to achieve a real-time speed of over 29, 36 and 45 high FPS; 32.9dB/0.889, 31.86dB/0.801 and 30.94dB/0.728 PSNR/SSIM values for 2x, 3x and 4x scaling factors respectively.

Abstract

The ramp generator is a crucial circuit component in the design of switching power supply, analog to digital converters, CMOS image sensors, and many other essential circuits. Conventionally, the ramp generator has been implemented with a CMOS comparator that works on the hysteresis principle, which leads to an inherent systemic delay. This paper aims at improving the overall circuit speed by proposing the design of a ramp generator using CMOS proteretic comparator in 180nm TSMC process. It is established from the post-layout simulation that a speedup of 25% is achieved by replacing a hysteretic comparator with a proteretic comparator, and this is done with a trade-off in power consumption and circuit area.

Abstract

The effect of Residual Mechanical Jitter (RMJ) has been analysed on the incoherent beam combination through various atmospheric turbulences. It is observed that the RMJ effect is more prominent for lower turbulence than higher turbulence.

Abstract

In this study, we characterize biosamples using a microstrip patch based Radio Frequency (RF) sensor by integrating both resonant and non-resonant parameters. In particular, the resonance frequency, the resonance amplitude, −10 dB bandwidth, range of operation, and return loss at center frequency were employed to perform the analysis. The idea was to utilize multiple parameters to build a comprehensive understanding of different bio-samples. In order to demonstrate the working of the methodology, two sets of analytes were studied-(i) aqueous solutions of biomolecules and (ii) body fluids. For both sets of analytes, the method demonstrated significant efficiency in performing qualitative characterization.

Abstract

In this article, we present a novel target classification technique by mmWave frequency modulated continuous wave (FMCW) Radars using the Machine Learning on raw data features obtained from range fast Fourier transform (FFT) plot. FFT plots are extracted from the measured raw data obtained with a Radar operating in the frequency range of 77 - 81 GHz. The features such as peak, width, area, standard deviation, and range on range FFT plot peaks are extracted and fed to a machine learning model. Two light weight classification models such as Logistic Regression, Naive Bayes are explored to assess the performance. Based on the results, we demonstrate and achieve an accuracy of 86.9% using Logistic Regression. The proposed technique will be highly useful for several applications in cost-effective and reliable ground station traffic management systems for autonomous systems. The end-to-end framework presented here, expands the capabilities of mmWave Radar beyond range detection to classification. The implications of this added functionalities will facilitate utilization of mmWave Radars in computer vision, object recognition, and towards fully autonomous traffic control and management systems.