Synthesizing database programs for schema refactoring

Wang, Yuepeng; Dong, James; Shah, Rushi; Dillig, Işıl

doi:10.1145/3314221.3314588

Cited by 34 publications

(18 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Beyond checking feasibility, some deduction techniques used for synthesis can also provide additional information [18,32,54]. In particular, given a partial program P that is infeasible with respect to specification φ, several deduction engines can generate a set of other infeasible partial programs P 1 , .…”

Section: Definition 2 (Feasible Partial Program)mentioning

confidence: 99%

“…However, in the context of program synthesis, deductive reasoning can also provide feedback about states that have not been explored. For example, given a partial program that is infeasible, one can analyze the root cause of failure to infer other infeasible programs [18,54]. To deal with this difficulty, we propose an off-policy reinforcement learning algorithm that can improve the policy based on such additional feedback from the deduction engine.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Program Synthesis Using Deduction-Guided Reinforcement Learning

Chen

Wang

Bastani

et al. 2020

Computer Aided Verification

Self Cite

View full text Add to dashboard Cite

In this paper, we present a new program synthesis algorithm based on reinforcement learning. Given an initial policy (i.e. statistical model) trained off-line, our method uses this policy to guide its search and gradually improves it by leveraging feedback obtained from a deductive reasoning engine. Specifically, we formulate program synthesis as a reinforcement learning problem and propose a new variant of the policy gradient algorithm that can incorporate feedback from a deduction engine into the underlying statistical model. The benefit of this approach is twofold: First, it combines the power of deductive and statistical reasoning in a unified framework. Second, it leverages deduction not only to prune the search space but also to guide search. We have implemented the proposed approach in a tool called Concord and experimentally evaluate it on synthesis tasks studied in prior work. Our comparison against several baselines and two existing synthesis tools shows the advantages of our proposed approach. In particular, Concord solves 15% more benchmarks compared to Neo, a state-of-the-art synthesis tool, while improving synthesis time by 8.71× on benchmarks that can be solved by both tools.

show abstract

Section: Definition 2 (Feasible Partial Program)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Program Synthesis Using Deduction-Guided Reinforcement Learning

Chen

Wang

Bastani

et al. 2020

Computer Aided Verification

Self Cite

View full text Add to dashboard Cite

show abstract

“…Given this DSL, we have developed a regex synthesizer that performs top-down enumerative search over the space of possible regexes defined by the DSL. This is a standard approach adopted by many existing synthesis techniques [16,15,63,62]. Our regex synthesizer takes both positive and negative examples: a positive example is a string that should be accepted by the desired regex whereas a negative example is a string that should be rejected by the regex.…”

Section: The Synthesis Algorithmmentioning

confidence: 99%

“…Program synthesis has been increasingly applied to many application domains, e.g., software development tools [15,50,62,29], human-robot interaction [47,35], and end-user programming [32,22,36,6,38]. Many program synthesis techniques adopt a programming-by-example (PBE) paradigm, where users provide a set of input-output examples to describe the intended program behavior.…”

Section: Introductionmentioning

confidence: 99%

Interactive Program Synthesis by Augmented Examples

Zhang

Lowmanstone

Wang

et al. 2020

Proceedings of the 33rd Annual ACM Symposium on User Interface Software and Technology

View full text Add to dashboard Cite

Mark parts of an input as literal or general 2 Annotate parts of synthesized programs to be desired or undesired 3 View other similar examples and corner cases 4 Figure 1: A user interactively synthesizes a regular expression (regex) that accepts phone numbers starting with an optional '+'. After 1) adding some positive and negative examples, the user 2) marks the + sign as literal and the following numbers as a general class of digits to specify how individual characters should be treated by the synthesizer. To refine the synthesis result, the user can either 3) directly mark subexpressions as desired or undesired in the final regex, or 4) ask for additional examples to validate and enhance her understanding of a regex candidate as well as identify and label counterexamples, therefore adding them to the set of user-provided input-output examples.

show abstract

“…Wang et al propose a technique [Wang et al 2019] for synthesizing equivalent database queries for an application that has undergone schema migration. They establish equivalence predicates across two versions of the schema by equating the columns and then employ sketch based synthesis for generating equivalent queries.…”

Section: Related Workmentioning

confidence: 99%

Synthesizing replacement classes

Samak

Kim

Rinard

2019

Proc. ACM Program. Lang.

View full text Add to dashboard Cite

We present a new technique for automatically synthesizing replacement classes. The technique starts with an original class O and a potential replacement class R, then uses R to synthesize a new class that implements the same interface and provides the same functionality as O. Critically, our technique works with a synthesized inter-class equivalence predicate between the states of O and R. It uses this predicate to ensure that original and synthesized methods leave corresponding O and R objects in equivalent states. The predicate therefore enables the technique to synthesize individual replacement methods in isolation while still obtaining a replacement class that leaves the original and replacement objects in equivalent states after arbitrarily long method invocation sequences. We have implemented the technique as part of a tool, named M , and evaluated it using open-source Java classes. The results highlight the e ectiveness of M in synthesizing replacement classes. CCS Concepts: • Software and its engineering → Programming by example; Object oriented frameworks.

show abstract

Synthesizing database programs for schema refactoring

Cited by 34 publications

References 61 publications

Program Synthesis Using Deduction-Guided Reinforcement Learning

Program Synthesis Using Deduction-Guided Reinforcement Learning

Interactive Program Synthesis by Augmented Examples

Synthesizing replacement classes

Contact Info

Product

Resources

About