Phys: probabilistic physical unit assignment and inconsistency detection

Kate, Sayali; Ore, John-Paul; Zhang, Xiangyu; Elbaum, Sebastian; Xu, Zhaogui

doi:10.1145/3236024.3236035

Cited by 18 publications

(15 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, they automatically refine string variables into some that represent passwords, some that represents file system paths, etc. Similarly, Kate et al [37] use probabilistic inference to predict physical units in scientific code. While they use probabilistic reasoning to perform this task, no learning is employed.…”

Section: Introductionmentioning

confidence: 99%

Typilus: neural type hints

Allamanis

Barr

Ducousso

et al. 2020

Proceedings of the 41st ACM SIGPLAN Conference on Programming Language Design and Implementation

100

View full text Add to dashboard Cite

Type inference over partial contexts in dynamically typed languages is challenging. In this work, we present a graph neural network model that predicts types by probabilistically reasoning over a program's structure, names, and patterns. The network uses deep similarity learning to learn a TypeSpacea continuous relaxation of the discrete space of types-and how to embed the type properties of a symbol (i.e. identifier) into it. Importantly, our model can employ one-shot learning to predict an open vocabulary of types, including rare and user-defined ones. We realise our approach in Typilus for Python that combines the TypeSpace with an optional type checker. We show that Typilus accurately predicts types. Typilus confidently predicts types for 70% of all annotatable symbols; when it predicts a type, that type optionally type checks 95% of the time. Typilus can also find incorrect type annotations; two important and popular open source libraries, fairseq and allennlp, accepted our pull requests that fixed the annotation errors Typilus discovered. CCS Concepts: • Computing methodologies → Machine learning; • Software and its engineering → Language features.

show abstract

Section: Introductionmentioning

confidence: 99%

Typilus: neural type hints

Allamanis

Barr

Ducousso

et al. 2020

Proceedings of the 41st ACM SIGPLAN Conference on Programming Language Design and Implementation

100

View full text Add to dashboard Cite

show abstract

“…In practice, existing literature of probability inference typically makes use of pre-defined prior probability values derived from domain knowledge [84], [45], [36], [58], [21], [60], [50].…”

Section: A Random Variables and Probabilistic Constraintsmentioning

confidence: 99%

“…Dietz et al also leverage probabilistic inference to localize source code bugs [36]. Besides, probabilistic techniques are widely used for binary analysis [87], [61], physical unit security [45], program enhancement [49], and vulnerability detection [36], [58]. To the best of our knowledge, NETPLIER is the first approach that enforces probabilistic analysis on protocol reverse engineering.…”

Section: Related Workmentioning

confidence: 99%

NetPlier: Probabilistic Network Protocol Reverse Engineering from Message Traces

Zhang

Wang

et al. 2021

Proceedings 2021 Network and Distributed System Security Symposium

Self Cite

View full text Add to dashboard Cite

Network protocol reverse engineering is an important challenge with many security applications. A popular kind of method leverages network message traces. These methods rely on pair-wise sequence alignment and/or tokenization. They have various limitations such as difficulties of handling a large number of messages and dealing with inherent uncertainty. In this paper, we propose a novel probabilistic method for network trace based protocol reverse engineering. It first makes use of multiple sequence alignment to align all messages and then reduces the problem to identifying the keyword field from the set of aligned fields. The keyword field determines the type of a message. The identification is probabilistic, using random variables to indicate the likelihood of each field (being the true keyword). A joint distribution is constructed among the random variables and the observations of the messages. Probabilistic inference is then performed to determine the most likely keyword field, which allows messages to be properly clustered by their true types and enables the recovery of message format and state machine. Our evaluation on 10 protocols shows that our technique substantially outperforms the state-of-the-art and our case studies show the unique advantages of our technique in IoT protocol reverse engineering and malware analysis.

show abstract

“…VI. Re l a t e d W o r k a) Name-based Program Analysis: Various analyses exploit the rich information provided by identifier names, e.g., to find bugs [2]- [5] and vulnerabilities [46], to mine specifications [6 ], to infer types based on identifier names as implicit type hints [7], [8 ], to predict the name of a method [9], to complete partial code using a learned language model [1 0 ], to identify inappropriate names [1 1 ], to suggest more suitable names [1 2 ], to resolve fully qualified type names of methods, variables, etc. in a given code snippet [47], or to map APIs between programming languages based on an embedding of code tokens [18].…”

Section: B Threats To External Validitymentioning

confidence: 99%

IdBench: Evaluating Semantic Representations of Identifier Names in Source Code

Wainakh

Rauf

Pradel

2021

2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE)

View full text Add to dashboard Cite

Identifier names convey useful information about the intended semantics of code. Name-based program analyses use this information, e.g., to detect bugs, to predict types, and to improve the readability of code. At the core of namebased analyses are semantic representations of identifiers, e.g., in the form of learned embeddings. The high-level goal of such a representation is to encode whether two identifiers, e.g., le n and s i z e , are semantically similar. Unfortunately, it is currently unclear to what extent semantic representations match the semantic relatedness and similarity perceived by developers. This paper presents IdBench, the first benchmark for evaluating semantic representations against a ground truth created from thousands of ratings by 500 software developers. We use IdBench to study state-of-the-art embedding techniques proposed for natural language, an embedding technique specifically designed for source code, and lexical string distance functions. Our results show that the effectiveness of semantic representations varies significantly and that the best available embeddings successfully represent semantic relatedness. On the downside, no existing technique provides a satisfactory representation of semantic similarities, among other reasons because identifiers with opposing meanings are incorrectly considered to be similar, which may lead to fatal mistakes, e.g., in a refactoring tool. Studying the strengths and weaknesses of the different techniques shows that they complement each other. As a first step toward exploiting this complementarity, we present an ensemble model that combines existing techniques and that clearly outperforms the best available semantic representation.Index Terms-source code, neural networks, embeddings, identifiers, benchmark I. In t r o d u c t io n Identifier names play an important role in writing, understanding, and maintaining high-quality source code [1]. Because they convey information about the meaning of variables, functions, classes, and other program elements, developers often rely on identifiers to understand code written by themselves and others. Beyond developers, various automated techniques analyze, use, and improve identifier names. For example, identifiers have been used to find programming errors [2 ]-[5], to mine specifications [6 ], to infer types [7], [8 ], to predict the name of a method [9], or to complete partial code using a learned language model [10]. Techniques for

show abstract

Phys: probabilistic physical unit assignment and inconsistency detection

Cited by 18 publications

References 17 publications

Typilus: neural type hints

Typilus: neural type hints

NetPlier: Probabilistic Network Protocol Reverse Engineering from Message Traces

IdBench: Evaluating Semantic Representations of Identifier Names in Source Code

Contact Info

Product

Resources

About