MultiFix: Learning to Repair Multiple Errors by Optimal Alignment Learning

We consider the problem of learning to repair erroneous C programs by learning optimal alignments with correct programs. Since the previous approaches fix a single error in a line, it is inevitable to iterate the fixing process until no errors remain. In this work, we propose a novel sequence-to-sequence learning framework for fixing multiple program errors at a time. We introduce the edit-distance-based data labeling approach for program error correction. Instead of labeling a program repair example by pairing an erroneous program with a line fix, we label the example by paring an erroneous program with an optimal alignment to the corresponding correct program produced by the edit-distance computation. We evaluate our proposed approach on a publicly available dataset (DeepFix dataset) that consists of erroneous C programs submitted by novice programming students. On a set of 6,975 erroneous C programs from the DeepFix dataset, our approach achieves the state-of-the-art result in terms of full repair rate on the DeepFix dataset (without extra data such as compiler error message or additional source codes for pre-training).

Using Optimal Transport as Alignment Objective for fine-tuning Multilingual Contextualized Embeddings

Recent studies have proposed different methods to improve multilingual word representations in contextualized settings including techniques that align between source and target embedding spaces. For contextualized embeddings, alignment becomes more complex as we additionally take context into consideration. In this work, we propose using Optimal Transport (OT) as an alignment objective during fine-tuning to further improve multilingual contextualized representations for downstream cross-lingual transfer. This approach does not require word-alignment pairs prior to fine-tuning that may lead to sub-optimal matching and instead learns the word alignments within context in an unsupervised manner. It also allows different types of mappings due to soft matching between source and target sentences. We benchmark our proposed method on two tasks (XNLI and XQuAD) and achieve improvements over baselines as well as competitive results compared to similar recent works.

HOTTER: Hierarchical Optimal Topic Transport with Explanatory Context Representations

Natural language processing (NLP) is often the backbone of today's systems for user interactions, information retrieval and others. Many of such NLP applications rely on specialized learned representations (e.g. neural word embeddings, topic models) that improve the ability to reason about the relationships between documents of a corpus. Paired with the progress in learned representations, the similarity metrics used to compare representations of documents are also evolving, with numerous proposals differing in computation time or interpretability. In this paper we propose an extension to a specific emerging hybrid document distance metric which combines topic models and word embeddings: the Hierarchical Optimal Topic Transport (HOTT). In specific, we extend HOTT by using context-enhanced word representations. We provide a validation of our approach on public datasets, using the language model BERT for a document categorization task. Results indicate competitive performance of the extended HOTT metric. We furthermore apply the HOTT metric and its extension to support educational media research, with a retrieval task of matching topics in German curricula to educational textbooks passages, along with offering an auxiliary explanatory document representing the dominant topic of the retrieved document. In a user study, our explanation method is preferred over regular topic keywords.

Optimal Word Segmentation for Neural Machine Translation into Dravidian Languages

Dravidian languages, such as Kannada and Tamil, are notoriously difficult to translate by state-of-the-art neural models. This stems from the fact that these languages are morphologically very rich as well as being low-resourced. In this paper, we focus on subword segmentation and evaluate Linguistically Motivated Vocabulary Reduction (LMVR) against the more commonly used SentencePiece (SP) for the task of translating from English into four different Dravidian languages. Additionally we investigate the optimal subword vocabulary size for each language. We find that SP is the overall best choice for segmentation, and that larger dictionary sizes lead to higher translation quality.

Generating An Optimal Interview Question Plan Using A Knowledge Graph And Integer Linear Programming

Given the diversity of the candidates and complexity of job requirements, and since interviewing is an inherently subjective process, it is an important task to ensure consistent, uniform, efficient and objective interviews that result in high quality recruitment. We propose an interview assistant system to automatically, and in an objective manner, select an optimal set of technical questions (from question banks) personalized for a candidate. This set can help a human interviewer to plan for an upcoming interview of that candidate. We formalize the problem of selecting a set of questions as an integer linear programming problem and use standard solvers to get a solution. We use knowledge graph as background knowledge in this formulation, and derive our objective functions and constraints from it. We use candidate's resume to personalize the selection of questions. We propose an intrinsic evaluation to compare a set of suggested questions with actually asked questions. We also use expert interviewers to comparatively evaluate our approach with a set of reasonable baselines.

Optimization of the Design of Hexapod Platform

The main objective of this research is to present a study on the design optimization of the 6-RUS Stewart platform. The geometric and kinematic models are calculated and the singular positions are determined, then its translation and orientation workspace are determining. The direct geometric model of the studied platform was determined by using a proposed hybrid method.

Stochastic mathematical models

The purpose of this research is to study and create a stochastic mathematical model based on a renewable energy source (wind). The question of finding optimal values for the variables of the mathematical model subject to stochastic conditions is one of the random mathematical problems, which require special stochastic methods to solve in the general case.

An Enhanced Real-Time Routing Protocol with load distributionERTLD ;Compared to Baseline Routing Protocols In MWSN

Mobile wireless sensor network (MWSN) is a wireless ad hoc network that consists of avery large number of tiny sensor nodes communicating with each other in which sensornodes are either equipped with motors for active mobility or attached to mobile objectsfor passive mobility. A real-time routing protocol for MWSN is an exciting area of research because messages in the network are delivered according to their end-to-end deadlines (packet lifetime) while sensor nodes are mobile. This paper proposes an enhanced realtime with load distribution (ERTLD) routing protocol for MWSN which is based on our previousrouting protocol RTLD. ERTLD utilized corona mechanism and optimal forwardingmetrics to forward the data packet in MWSN. It computes the optimal forwarding nodebased on RSSI, remaining battery level of sensor nodes and packet delayover one-hop. ERTLDensures high packet delivery ratio and experiences minimum end-to-end delay in WSNand MWSN compared to baseline routing protocol. . In this paper we consider a highly dynamic wireless sensor network system in which the sensor nodes and the base station(sink) are mobile.ERTLD has been studied and verified and compared with baseline routing protocols RTLD,MM-SPEED , RTLCthrough Network Simulator- 2(NS2)

Potential of Free and Open Source Geographic Information Systems (FOSS_GIS) Applications in Optimal Site Selection Analysis of Urban Project

Free and Open Source software (FOSS) is one of computer software, which source code can be accessed, freely used, modify, and distribute by anyone. It is produced by many of people or organizations, and distributed under licenses that comply with the open source definition. This software has recently begun to play an important role in the academic and scientific research field, as in the professional field. In the past few decades, Geographic Information Systems (GIS) has seen very high growth rate, and this development included each of commercial and open source GIS software. This research aims to show the great potential of Free and Open Source Geographic Information Systems (FOSS_GIS), and motivating to adopt it in developing countries, as a means to reduce licensing costs, promote local technological development through access to the source code and developing these systems. A case study is taken, in which we have tried to highlight the most important advantages of this software (i.e. FOSS_GIS), such as ease of implementation and good use, the ability to analyze and display of spatial data, professional maps production, and functionality emulator to commercial GIS software. The case study included the methodology of spatial suitability analysis, which is one of the main tasks of GIS; this methodology has been applied to choose the optimal site of urban project in Sheikh Badr area (Tartous Governorate) by proposing a set of general conditions. Using optimal site selection analysis of urban project helps to avoid the indiscriminate expansion and the irregular land-use. The free and open source software QGIS was used in this research, as well as the algorithms and tools of GRASS and SAGA softwares. Key Words: FOSS_GIS, spatial analysis, QGIS, optimal site selection, Sheikh Bader area

Developing an Approach for Data Steganography in Image by Segmentation to Different Size Random Blocks

In this paper we propose a new steganography algorithm by a random image segmentation to blocks of different size with several steganographic algorithms to increasing the security of hiding data. Both color and gray images have been used as cover files for our method.