Active Learning of Points-To Specifications

نویسندگان

  • Osbert Bastani
  • Rahul Sharma
  • Alexander Aiken
  • Percy Liang
چکیده

Large libraries pose signi€cant challenges to static points-to analysis. A popular solution is to have a human analyst provide points-to speci€cations that summarize relevant behaviors of library code, which can substantially improve precision and scalability, and furthermore handle missing code such as native code. We propose Atlas, a tool that automatically infers points-to speci€cations. Atlas synthesizes test cases that exercise the library code, and then generates points-to speci€cations based on observations from these executions. Atlas automatically infers 97% of speci€cations for the Java Collections API, and discovers 20% more points-to edges than existing, handwriŠen speci€cations.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

High-Dimensional Unsupervised Active Learning Method

In this work, a hierarchical ensemble of projected clustering algorithm for high-dimensional data is proposed. The basic concept of the algorithm is based on the active learning method (ALM) which is a fuzzy learning scheme, inspired by some behavioral features of human brain functionality. High-dimensional unsupervised active learning method (HUALM) is a clustering algorithm which blurs the da...

متن کامل

دیدگاه دانشجویان پزشکی پیرامون تجربهی مشارکت دانشجویی در فرآیند یاددهی ـ یادگیری درس پوست به دو شیوهی بحث گروهی و سخنرانی

Background and Aim: The World Federation for Medical Education developed standards for medical education and recommended medical schools to encourage their students to take active responsibility for their self-learning, and prepare them for lifelong learning. The purpose of this study was to compare the viewpoints of medical students about the effectiveness of their participation in two methods...

متن کامل

Information-Based Objective Functions for Active Data Selection

Learning can be made more efficient if we can actively select particularly salient data points. Within a Bayesian learning framework, objective functions are discussed that measure the expected informativeness of candidate measurements. Three alternative specifications of what we want to gain information about lead to three different criteria for data selection. All these criteria depend on the...

متن کامل

COALA - Correlation-Aware Active Learning of Link Specifications

Link Discovery plays a central role in the creation of knowledge bases that abide by the five Linked Data principles. Over the last years, several active learning approaches have been developed and used to facilitate the supervised learning of link specifications. Yet so far, these approaches have not taken the correlation between unlabeled examples into account when requiring labels from their...

متن کامل

بررسی شاخص های عملکردی بیمارستان های آموزشی، عمومی دانشگاه علوم پزشکی ایران

The present study is the result of a comprehensive and continuous study for assessing the performance indexes in public-educational hospitals of Iran University of Medical Sciences. Considering disclosed standards of Ministry of health and medical education, five public-educational hospitals in a descriptive-analytical cross-sectional study have been investigated with a comparative perspective,...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:
  • CoRR

دوره abs/1711.03239  شماره 

صفحات  -

تاریخ انتشار 2017