Date: 2019-11-26T18:44:44+01:00 Unread emails: 64 Paper titles: 154 Uniq paper titles: 101
- Generating precise error specifications for C: a zero shot learning approach (4)
Abstract: In C programs, error specifications, which specify the value range that each fun ...
ction returns to indicate failures, are widely used to check and propagate errors for the sake of reliability and security. Various kinds of C analyzers employ error … - Improving Bug Detection via Context-Based Code Representation Learning and Attention-Based Neural Networks (4)
Abstract: • Rule-based bug detection. In this type of approaches, several programming ru ...
les are predefined to statically detect common programming flaws or defects. A popular example of this type of approaches is FindBugs [Hovemeyer and Pugh 2007]. While … - Learning to Fuzz from Symbolic Execution with Application to Smart Contracts (4)
Abstract: Fuzzing and symbolic execution are two complementary techniques for discovering ...
software vulnerabilities. Fuzzing is fast and scalable, but can be ineffective when it fails to randomly select the right inputs. Symbolic execution is thorough but slow and … - Commit2Vec: Learning Distributed Representations of Code Changes (4)
Abstract: Deep learning methods, which have found successful applications in fields like i ...
mage classification and natural language processing, have recently been applied to source code analysis too, due to the enormous amount of freely available source … - Evaluating Semantic Representations of Source Code (4)
Abstract: Learned representations of source code enable various software developer tools, ...
eg, to detect bugs or to predict program properties. At the core of code representations often are word embeddings of identifier names in source code, because identifiers … - AutoPandas: Neural-Backed Generators for Program Synthesis (3)
Abstract: Developers nowadays have to contend with a growing number of APIs. Many of these ...
APIs are very useful to developers, increasing the ease of code re-use. API functions provide implementations of functionalities that are often more efficient … - Do People Prefer" Natural" code? (3)
Abstract: Natural code is known to be very repetitive (much more so than natural language ...
corpora); furthermore, this repetitiveness persists, even after accounting for the simpler syntax of code. However, programming languages are very expressive … - MACHINE LEARNING FOR CODE SYNTHESIS AND ANALYSIS (3)
Abstract: Deep learning has been successfully applied to a wide array of problems due to i ...
ts versatility and the variety of algorithms being developed. Deep learning for natural language processing makes use of recurrent networks, attention layers, copy … - Exploiting Token and Path-based Representations of Code for Identifying Security-Relevant Commits (3)
Abstract: Public vulnerability databases such as CVE and NVD account for only 60% of secur ...
ity vulnerabilities present in open-source projects, and are known to suffer from inconsistent quality. Over the last two years, there has been considerable growth in … - CodeSearchNet Challenge: Evaluating the State of Semantic Code Search (3)
Abstract: Semantic code search is the task of retrieving relevant code given a natural lan ...
guage query. While related to other information retrieval tasks, it requires bridging the gap between the language used in code (often abbreviated and highly technical) … - Modeling Security Weaknesses to Enable Practical Run-time Defenses (3)
Abstract: Security weaknesses are sometimes caused by patterns in human behaviors. However ...
, it can be difficult to identify such patterns in a practical, yet accurate way. In order to fix security weaknesses, it is crucial to identify them. Useful systems to … - Encodings for Enumeration-Based Program Synthesis (3)
Abstract: Program synthesis is the problem of finding a program that satisfies a given spe ...
cification. Most program synthesizers are based on enumerating program candidates that satisfy the specification. Recently, several new tools for program … - N-Grams as a Measure of Naturalness and Complexity (3)
Abstract: We live in a time where software is used everywhere. It is used even for creatin ...
g other software by helping developers with writing or generating new code. To do this properly, metrics to measure software quality are being used to evaluate the final … - Relational Verification using Reinforcement Learning (3)
Abstract: Authors' addresses: Jia Chen, Department of Computer Science, University of Texa ...
s at Austin, Austin, Texas, 78712-0233, USA, jchen@ cs. utexas. edu; Jiayi Wei, Department of Computer Science, University of Texas at Austin, Austin, Texas, 78712 … - Improve Language Modelling for Code Completion through Statement Level Language Model based on Statement Embedding Generated by BiLSTM (3)
Abstract: Language models such as RNN, LSTM or other variants have been widely used as gen ...
erative models in natural language processing. In last few years, taking source code as natural languages, parsing source code into a token sequence and using a … - A machine learning based automatic folding of dynamically typed languages (3)
Abstract: The popularity of dynamically typed languages has been growing strongly lately. ...
Elegant syntax of such languages like javascript, python, PHP and ruby pays back when it comes to finding bugs in large codebases. The analysis is hindered by … - Structural Language Models for Any-Code Generation (3)
Abstract: We address the problem of Any-Code Generation (AnyGen)-generating code without a ...
ny restriction on the vocabulary or structure. The state-of-the-art in this problem is the sequence-to-sequence (seq2seq) approach, which treats code as a sequence … - zkay: Specifying and Enforcing Data Privacy in Smart Contracts (2)
Abstract: Privacy concerns of smart contracts are a major roadblock preventing their wider ...
adoption. A promising approach to protect private data is hiding it with cryptographic primitives and then enforcing correctness of state updates by Non-Interactive Zero … - Evaluating Lexical Approximation of Program Dependence (2)
Abstract: Complex dependence analysis typically provides an underpinning approximation of ...
true program dependence. We investigate the effectiveness of using lexical information to approximate such dependence, introducing two new deletion … - Learning based Methods for Code Runtime Complexity Prediction (2)
Abstract: Predicting the runtime complexity of a programming code is an arduous task. In f ...
act, even for humans, it requires a subtle analysis and comprehensive knowledge of algorithms to predict time complexity with high fidelity, given any code. As per … - Assessing the Generalizability of code2vec Token Embeddings (2)
Abstract: Many Natural Language Processing (NLP) tasks, such as sentiment analysis or synt ...
actic parsing, have benefited from the development of word embedding models. In particular, regardless of the training algorithms, the learned embeddings have … - Neural Program Synthesis By Self-Learning (2)
Abstract: Neural inductive program synthesis is a task generating instructions that can pr ...
oduce desired outputs from given inputs. In this paper, we focus on the generation of a chunk of assembly code that can be executed to match a state change inside the … - Universal Approximation with Certified Networks (2)
Abstract: Training neural networks to be certifiably robust is a powerful defense against ...
adversarial attacks. However, while promising, state-of-the-art results with certified training are far from satisfactory. Currently, it is very difficult to train a neural network … - DeepVS: An Efficient and Generic Approach for Source Code Modeling Usage (2)
Abstract: Recently deep learning-based approaches have shown great potential in the modeli ...
ng of source code for various software engineering tasks. These techniques lack adequate generalization and resistance to acclimate the use of such models in a … - Deep Transfer Learning for Source Code Modeling (2)
Abstract: In recent years, deep learning models have shown great potential in source code ...
modeling and analysis. Generally, deep learning-based approaches are problem-specific and data-hungry. A challenging issue of these approaches is that they … - Twin-Finder: Integrated Reasoning Engine for Pointer-related Code Clone Detection (2)
Abstract: Detecting code clones is crucial in various software engineering tasks. In parti ...
cular, code clone detection can have significant uses in the context of analyzing and fixing bugs in large scale applications. However, prior works, such as machine learning … - Learning Lenient Parsing & Typing via Indirect Supervision (2)
Abstract: Both professional coders and teachers frequently deal with imperfect (fragmentar ...
y, incomplete, ill-formed) code. Such fragments are common in StackOverflow; students also frequently produce ill-formed code, for which instructors, TAs (or students … - Class Name Recommendation based on Graph Embedding of Program Elements (2)
Abstract: In software development, the quality of identifier names is important because it ...
greatly affects program comprehension for developers. However, naming identifiers that appropriately represent the nature or behavior of program elements such as … - A Deep Learning Model for Source Code Generation (2)
Abstract: ABSTRACT Natural Language Processing (NLP) models have been used extensively to ...
study relationship among words in a corpus. Inspired by models such as n-gram we developed a model for analyzing source code via its Abstract Syntax … - On The Quality of Identifiers in Test Code (2)
Abstract: Meaningful, expressive identifiers in source code can enhance the readability an ...
d reduce comprehension efforts. Over the past years, researchers have devoted considerable effort to understanding and improving the naming quality of identifiers … - Poster: Finding JavaScript Name Conflicts on the Web (2)
Abstract: Including JavaScript code from many different hosts is a popular practice in dev ...
eloping web applications. For example, to include a social plugin like the Facebook Like button, a web developer needs to only include a script from facebook … - Formal Verification of Workflow Policies for Smart Contracts in Azure Blockchain (1)
Abstract: Ensuring correctness of smart contracts is paramount to ensuring trust in blockc ...
hain-based systems. This paper studies the safety and security of smart contracts in the Azure Blockchain Workbench, an enterprise Blockchain-as-a-Service offering from … - NutBaaS: A Blockchain-as-a-Service Platform (1)
Abstract: Blockchain, originated from Bitcoin system, has drawn intense attention from the ...
academic community because of its decentralization, persistency, anonymity and auditability. In the past decade, the blockchain technology has evolved and became … - CLN2INV: Learning Loop Invariants with Continuous Logic Networks (1)
Abstract: Program verification offers a framework for ensuring program correctness and the ...
refore systematically eliminating different classes of bugs. Inferring loop invariants is one of the main challenges behind automated verification of real-world programs … - Word Embedding Algorithms as Generalized Low Rank Models and their Canonical Form (1)
Abstract: Word embedding algorithms produce very reliable feature representations of words ...
that are used by neural network models across a constantly growing multitude of NLP tasks. As such, it is imperative for NLP practitioners to understand how their … - Keep It Simple: Graph Autoencoders Without Graph Convolutional Networks (1)
Abstract: Graph autoencoders (AE) and variational autoencoders (VAE) recently emerged as p ...
owerful node embedding methods, with promising performances on challenging tasks such as link prediction and node clustering. Graph AE, VAE and most of their … - Disentangling Interpretable Generative Parameters of Random and Real-World Graphs (1)
Abstract: While a wide range of interpretable generative procedures for graphs exist, matc ...
hing observed graph topologies with such procedures and choices for its parameters remains an open problem. Devising generative models that closely reproduce real … - Towards Robust Direct Perception Networks for Automated Driving (1)
Abstract: We consider the problem of engineering robust direct perception neural networks ...
with output being regression. Such networks take high dimensional input image data, and they produce affordances such as the curvature of the upcoming road segment … - Static Detection of Event-Driven Races in HTML5-Based Mobile Apps (1)
Abstract: HTML5-based mobile apps are developed using standard web technologies such as HT ...
ML5, CSS, JavaScript, so they may also face with event-based races as web apps. The races in such mobile apps can be caused by various sources of … - EXTRACTION OF STATIC FEATURES FROM BINARY APPLICATIONS FOR MALWARE ANALYSIS (1)
Abstract: ...
- An Evalutation of Programming Language Models' performance on Software Defect Detection (1)
Abstract: This dissertation presents an evaluation of several language models on software ...
defect datasets. A language Model (LM)" can provide word representation and probability indication of word sequences as the core component of an NLP system." … - Enabling Efficient Parallelism for Applications with Dependences and Irregular Memory Accesses (1)
Abstract: Page 1. Enabling Efficient Parallelism for Applications with Dependences and Irr ...
egularMemory Accesses Dissertation Presented in Partial Fulfillment of the Requirements forthe Degree Doctor of Philosophy in the Graduate School of The Ohio State University … - Efficient Graph Generation with Graph Recurrent Attention Networks (1)
Abstract: We propose a new family of efficient and expressive deep generative models of gr ...
aphs, called Graph Recurrent Attention Networks (GRANs). Our model generates graphs one block of nodes and associated edges at a time. The block size and … - Reducing Features to Improve Link Prediction Performance in Location Based Social Networks, Non-Monotonically Selected Subset from Feature Clusters (1)
Abstract: In most cases, feature sets available for machine learning algorithms require a ...
feature engineering approach to pick the subset for optimal performance. During our link prediction research, we had observed the same challenge for features of … - Graph Enhanced Cross-Domain Text-to-SQL Generation (1)
Abstract: Semantic parsing is a fundamental problem in natural language understanding, as ...
it involves the mapping of natural language to structured forms such as executable queries or logic-like knowledge representations. Existing deep learning approaches … - Learning from Examples to Find Fully Qualified Names of API Elements in Code Snippets (1)
Abstract: Developers often reuse code snippets from online forums, such as Stack Overflow, ...
to learn API usages of software frameworks or libraries. These code snippets often contain ambiguous undeclared external references. Such external references make … - Improving Textual Network Learning with Variational Homophilic Embeddings (1)
Abstract: The performance of many network learning applications crucially hinges on the su ...
ccess of network embedding algorithms, which aim to encode rich network information into low-dimensional vertex-based vector representations. This paper … - Why is Developing Machine Learning Applications Challenging? A Study on Stack Overflow Posts (1)
Abstract: As smart and automated applications pervade our lives, an increasing number of s ...
oftware developers are required to incorporate machine learning (ML) techniques into application development. However, acquiring the ML skill set can be nontrivial … - A Survey of Compiler Testing (1)
Abstract: Compilers are important tools because they are a central piece of infrastructure ...
for building other software. Virtually every program that runs on a computer, ranging from operating systems over web browsers to small scripts written by end-users, has … - NeuroVectorizer: End-to-End Vectorization with Deep Reinforcement Learning (1)
Abstract: One of the key challenges arising when compilers vectorize loops for today's SIM ...
D-compatible architectures is whether to vectorize and/or interleave. Then, the compiler has to determine how many instructions to pack together and how many loop … - Beyond the Single Neuron Convex Barrier for Neural Network Certification (1)
Abstract: We propose a new parametric framework, called k-ReLU, for computing precise and ...
scalable convex relaxations used to certify neural networks. The key idea is to approximate the output of multiple ReLUs in a layer jointly instead of separately. This … - Novel positional encodings to enable tree-based transformers (1)
Abstract: Neural models optimized for tree-based problems are of great value in tasks like ...
SQL query extraction and program synthesis. On sequence-structured data, transformers have been shown to learn relationships across arbitrary pairs of positions more … - Program Synthesis for Programmers (1)
Abstract: Recent years have seen great progress in automated synthesis techniques that can ...
automatically generate code based on some intent expressed by the user, but communicating this intent remains a major challenge. When the expressed intent is … - Combining Program Analysis and Statistical Language Model for Code Statement Completion (1)
Abstract: Automatic code completion helps improve developers' productivity in their progra ...
mming tasks. A program contains instructions expressed via code statements, which are considered as the basic units of program execution. In this paper, we … - Sequence Model Design for Code Completion in the Modern IDE (1)
Abstract: Code completion is a tremendously popular tool for coding assistance, implemente ...
d across a wide range of programming languages and environments. In An Empirical Investigation of Code Completion Usage by Professional Software Developers … - Software Engineering Meets Deep Learning: A Literature Review (1)
Abstract: Deep learning (DL) is being used nowadays in many traditional software engineeri ...
ng (SE) problems and tasks, such as software documentation, defect prediction, and software testing. However, since the renaissance of DL techniques is … - A multi-stage anomaly detection scheme for augmenting the security in IoT-enabled applications (1)
Abstract: The synergy between data security and high intensive computing has envisioned th ...
e way to robust anomaly detection schemes which in turn necessitates the need for efficient data analysis. Data clustering is one of the most important components of … - Recognizing lines of code violating company-specific coding guidelines using machine learning (1)
Abstract: Software developers in big and medium-size companies are working with millions o ...
f lines of code in their codebases. Assuring the quality of this code has shifted from simple defect management to proactive assurance of internal code quality. Although … - Neural Speech Translation using Lattice Transformations and Graph Networks (1)
Abstract: Speech translation systems usually follow a pipeline approach, using word lattic ...
es as an intermediate representation. However, previous work assume access to the original transcriptions used to train the ASR system, which can limit applicability in … - The Internet of Things and Machine Learning, Solutions for Urban Infrastructure Management (1)
Abstract: Urban infrastructure management requires the ability to reason about a large-sca ...
le complex system: What is the state of the system? How can it be compactly represented and quantified? How is the system likely to evolve? Reasoning calls for … - Memory Augmented Recursive Neural Networks (1)
Abstract: Recursive neural networks have shown an impressive performance for modeling comp ...
ositional data compared to their recurrent counterparts. Although recursive neural networks are better at capturing long range dependencies, their … - Multi-Modal Attention Network Learning for Semantic Source Code Retrieval (1)
Abstract: Code retrieval techniques and tools have been playing a key role in facilitating ...
software developers to retrieve existing code fragments from available open-source repositories given a user query. Despite the existing efforts in improving the … - Towards neural networks that provably know when they don't know (1)
Abstract: It has recently been shown that ReLU networks produce arbitrarily over-confident ...
predictions far away from the training data. Thus, ReLU networks do not know when they don't know. However, this is a highly important property in safety critical … - ASTToken2Vec: An Embedding Method for Neural Code Completion (1)
Abstract: Code completion systems help programmers to write code more efficiently and to r ...
educe typographical errors by automatically suggesting the code fragment that the programmers likely to write next. This work attempts to increase prediction … - Concealment of iris features based on artificial noises (1)
Abstract: Although iris recognition verification is considered to be the safest method of ...
biometric verification, studies have shown that iris features may be illegally used. To protect iris features and further improve the security of iris recognition and … - Augmented Example-based Synthesis using Relational Perturbation Properties (1)
Abstract: Authors' addresses: Shengwei An, Purdue University, USA, an93@ purdue. edu; Rish ...
abh Singh, Google Brain, USA, rising@ google. com; Sasa Misailovic, UIUC, USA, misailo@ illinois. edu; Roopsha Samanta, Purdue University, USA, roopsha … - Embedding Symbolic Knowledge into Deep Networks (1)
Abstract: In this work, we aim to leverage prior symbolic knowledge to improve the perform ...
ance of deep models. We propose a graph embedding network that projects propositional formulae (and assignments) onto a manifold via an augmented Graph … - CPC: automatically classifying and propagating natural language comments via program analysis (1)
Abstract: Modern software systems usually contain enormous code comments which provide abu ...
ndant information that have been lever-aged to help perform various software engineering tasks, such as bug detection, specification inference, and code … - Coding as another language: a pedagogical approach for teaching computer science in early childhood (1)
Abstract: Computer programming is an essential skill in the 21st century and new policies ...
and frameworks aim at preparing students for computer science-related professions. Today, the development of new interfaces and block-programming languages … - A Generative Model for Molecular Distance Geometry (1)
Abstract: Computing equilibrium states for many-body systems, such as molecules, is a long ...
-standing challenge. In the absence of methods for generating statistically independent samples, great computational effort is invested in simulating these … - Selective Monitoring Without Delay for Probabilistic System (1)
Abstract: Monitoring is an efficient tool to check correctness in runtime. A monitor obser ...
ves the output produced by a Markov Chain and decides if the run is correct or faulty. A selective monitor chooses the observed letters with a policy. It skips letters to reduce … - Transferring Java Comments Based on Program Static Analysis (1)
Abstract: In the process of software development and maintenance, code comments can help d ...
evelopers reduce the time of reading source code, and thus improve their work efficiency. For large Java software projects, comments tend to appear in front of the … - Exploring Robust Neural Methods in Inductive Program Synhthesis (1)
Abstract: ABSTRACT Inductive Program Synthesis (IPS) is an attractive goal for AI research ...
ers as it provides a solution to the problem of getting programs to write code. Recent work has shown that neural networks are efficient tools for improving the … - Combining Constraint Languages via Abstract Interpretation (1)
Abstract: Constraint programming initially aims to be a declarative paradigm, but its ques ...
t for efficiency is mainly achieved through the development of ad-hoc algorithms, which are encapsulated in global constraints. In this paper, we explore the idea of … - Testing Neural Program Analyzers (1)
Abstract: Deep neural networks have been increasingly used in software engineering and pro ...
gram analysis tasks. They usually take a program and make some predictions about it, eg, bug prediction. We call these models neural program analyzers. The … - A comparison of end-to-end models for long-form speech recognition (1)
Abstract: End-to-end automatic speech recognition (ASR) models, including both attention-b ...
ased models and the recurrent neural network transducer (RNN-T), have shown superior performance compared to conventional systems. However, previous studies … - Inverse‐QSPR for de novo design: a review (1)
Abstract: The use of computer tools to solve chemistry‐related problems–chemoinformati ...
cs–has given rise to a large number of publications. Among all chemoinformatics techniques, the use of statistically based approaches for property predictions has … - SoCodeCNN: Program Source Code for Visual CNN Classification Using Computer Vision Methodology (1)
Abstract: Automated feature extraction from program source-code such that proper computing ...
resources could be allocated to the program is very difficult given the current state of technology. Therefore, conventional methods call for skilled human intervention in … - Code Generation as a Dual Task of Code Summarization (1)
Abstract: Code summarization (CS) and code generation (CG) are two crucial tasks in the fi ...
eld of automatic software development. Various neural network-based approaches are proposed to solve these two tasks separately. However, there exists a specific … - Parallel Iterative Edit Models for Local Sequence Transduction (1)
Abstract: We present a Parallel Iterative Edit (PIE) model for the problem of local sequen ...
ce transduction arising in tasks like Grammatical error correction (GEC). Recent approaches are based on the popular encoder-decoder (ED) model for sequence to … - LSC: Online Auto-Update Smart Contracts for Fortifying Blockchain-Based Log Systems (1)
Abstract: Smart contracts allow verifiable operations to be executed in blockchains, bring ...
ing new possibilities for trust establishment in trustless scenarios. However, smart contracts are cumbersome when used as security mechanisms in security scenarios … - Neural Attribution for Semantic Bug-Localization in Student Programs (1)
Abstract: Providing feedback is an integral part of teaching. Most open online courses on ...
programming make use of automated grading systems to support programming assignments and give real-time feedback. These systems usually rely on test results … - Speech Recognition with Augmented Synthesized Speech (1)
Abstract: Recent success of the Tacotron speech synthesis architecture and its variants in ...
producing natural sounding multi-speaker synthesized speech has raised the exciting possibility of replacing expensive, manually transcribed, domain-specific … - Semantic Preserving Generative Adversarial Models (1)
Abstract: We introduce generative adversarial models in which the discriminator is replace ...
d by a calibrated (non-differentiable) classifier repeatedly enhanced by domain relevant features. The role of the classifier is to prove that the actual and generated data differ … - Zoea--Composable Inductive Programming Without Limits (1)
Abstract: Automatic generation of software from some form of specification has been a long ...
standing goal of computer science research. To date successful results have been reported for the production of relatively small programs. This paper presents Zoea … - Translationese as a Language in" Multilingual" NMT (1)
Abstract: Machine translation has an undesirable propensity to produce" translationese" ar ...
tifacts, which can lead to higher BLEU scores while being liked less by human raters. Motivated by this, we model translationese and original (ie natural) text as … - Systems and methods for generating and using dynamic and localized route-based environmental information (1)
Abstract: Methods, systems, and computer-readable media for generating localized environme ...
ntal information along a route of travel are generally described, for example, to notify users of travel conditions along a specific route and/or to generate … - Progressive Processing of System-Behavioral Query (1)
Abstract: System monitoring has recently emerged as an effective way to analyze and counte ...
r advanced cyber attacks. The monitoring data records a series of system events and provides a global view of system behaviors in an organization. Querying such data to … - Program Synthesis by Type-Guided Abstraction Refinement (1)
Abstract: We consider the problem of type-directed component based synthesis where, given ...
a set of (typed) components and a query type, the goal is to synthesize a term that inhabits the query. Classical approaches based on proof search in intuitionistic … - Imitation-Projected Programmatic Reinforcement Learning (1)
Abstract: We study the problem of programmatic reinforcement learning, in which policies a ...
re represented as short programs in a symbolic language. Such programmatic policies can be more interpretable, generalizable, and amenable to formal verification than … - Deep Representation Learning for Code Smells Detection using Variational Auto-Encoder (1)
Abstract: Detecting code smells is an important research problem in the software maintenan ...
ce. It assists the subsequent steps of the refactoring process so as to improve the quality of the software system. However, most of existing approaches … - Coda: An End-to-End Neural Program Decompiler (1)
Abstract: Reverse engineering of binary executables is a critical problem in the computer ...
security domain. On the one hand, malicious parties may recover interpretable source codes from the software products to gain commercial advantages. On the … - Why do they ask? An exploratory study of crowd discussions about Android application programming interface in stack overflow (1)
Abstract: Nowadays, more and more Android developers prefer to seek help from Q&A website ...
like Stack Overflow, despite the rich official documentation. Several researches have studied the limitations of the official application programming … - Mode Personalization in Trip-Based Transit Routing (1)
Abstract: We study the problem of finding bi-criteria Pareto optimal journeys in public tr ...
ansit networks. We extend the Trip-Based Public Transit Routing (TB) approach [Sascha Witt, 2015] to allow for users to select modes of interest at query time. As a first step … - Neural Relational Inference with Fast Modular Meta-learning (1)
Abstract: Graph neural networks (GNNs) are effective models for many dynamical systems con ...
sisting of entities and relations. Although most GNN applications assume a single type of entity and relation, many situations involve multiple types of interactions … - Abstraction Mechanism on Neural Machine Translation Models for Automated Program Repair (1)
Abstract: Bug fixing is a time-consuming task in software development. Automated bug repai ...
r tools are created to fix programs with little or no human effort. There are many existing tools based on the generate-and-validate (G&V) approach, which is an … - Online Robustness Training for Deep Reinforcement Learning (1)
Abstract: In deep reinforcement learning (RL), adversarial attacks can trick an agent into ...
unwanted states and disrupt training. We propose a system called Robust Student-DQN (RS-DQN), which permits online robustness training alongside Q networks … - A (CO) ALGEBRAIC APPROACH TO PROGRAMMING AND VERIFYING COMPUTER NETWORKS (1)
Abstract: As computer networks have grown into some of the most complex and critical compu ...
ting systems today, the means of configuring them have not kept up: they remain manual, low-level, and ad-hoc. This makes network operations expensive … - Compiler Auto-Vectorization with Imitation Learning (1)
Abstract: Modern microprocessors are equipped with single instruction multiple data (SIMD) ...
or vector instruction sets which allow compilers to exploit fine-grained data level parallelism. To exploit this parallelism, compilers employ auto-vectorization … - Méthodes d'apprentissage statistique pour le criblage virtuel de médicament Machine learning approaches for drug virtual screening (1)
Abstract: Cystic Fibrosis (CF) is the most common genetic disease among Caucasians, affect ...
ing approximately one in 3500 birth [193], and one person in 25 is an asymptomatic heterozygote carrier. It is an autosomal recessive genetic disease … - On the use of supervised machine learning for assessing schedulability: application to Ethernet TSN (1)
Abstract: In this work, we ask if Machine Learning (ML) can provide a viable alternative t ...
o conventional schedulability analysis to determine whether a real-time Ethernet network meets a set of timing constraints. Otherwise said, can an algorithm learn …