The three pillars of machine programming

Gottschlich, Justin; Solar-Lezama, Armando; Tatbul, Nesime; Carbin, Michael; Rinard, Martin; Barzilay, Regina; Amarasinghe, Saman; Tenenbaum, Joshua B.; Mattson, Tim

doi:10.1145/3211346.3211355

Cited by 41 publications

(29 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditional heuristics thus require re-tuning for each hardware configuration. Unlike traditional heuristics, LEA can adapt itself to the underlying hardware, the user's data, and the user's objective, all at the same time [2].…”

Section: Discussionmentioning

confidence: 99%

LEA: A Learned Encoding Advisor for Column Stores

Cen¹,

Kipf²,

Marcus³

et al. 2021

Fourth Workshop in Exploiting AI Techniques for Data Management

View full text Add to dashboard Cite

Data warehouses organize data in a columnar format to enable faster scans and better compression. Modern systems offer a variety of column encodings that can reduce storage footprint and improve query performance. Selecting a good encoding scheme for a particular column is an optimization problem that depends on the data, the query workload, and the underlying hardware.We introduce Learned Encoding Advisor (LEA), a learned approach to column encoding selection. LEA is trained on synthetic datasets with various distributions on the target system. Once trained, LEA uses sample data and statistics (such as cardinality) from the user's database to predict the optimal column encodings. LEA can optimize for encoded size, query performance, or a combination of the two. Compared to the heuristic-based encoding advisor of a commercial column store on TPC-H, LEA achieves 19% lower query latency while using 26% less space.

show abstract

Section: Discussionmentioning

confidence: 99%

LEA: A Learned Encoding Advisor for Column Stores

Cen¹,

Kipf²,

Marcus³

et al. 2021

Fourth Workshop in Exploiting AI Techniques for Data Management

View full text Add to dashboard Cite

show abstract

“…More broadly, applying reinforcement learning to systems problems can be viewed as machine programming (MP) [7], autonomously inventing new policies and adapting to new environments. However, MP extends well beyond applications of reinforcement learning to systems.…”

Section: Related Workmentioning

confidence: 99%

Learned garbage collection

Cen

Marcus

Mao

et al. 2020

Proceedings of the 4th ACM SIGPLAN International Workshop on Machine Learning and Programming Languages

Self Cite

View full text Add to dashboard Cite

Several programming languages use garbage collectors (GCs) to automatically manage memory for the programmer. Such collectors must decide when to look for unreachable objects to free, which can have a large performance impact on some applications. In this preliminary work, we propose a design for a learned garbage collector that autonomously learns over time when to perform collections. By using reinforcement learning, our design can incorporate user-defined reward functions, allowing an autonomous garbage collector to learn to optimize the exact metric the user desires (e.g., request latency or queries per second). We conduct an initial experimental study on a prototype, demonstrating that an approach based on tabular Q learning may be promising.

show abstract

“…Machine programming (MP), the automatic generation of software, is showing early signs of fundamentally transforming the way software is developed [15]. A key ingredient employed by MP is the deep neural network (DNN), which has emerged as an effective means to semi-autonomously implement many complex software systems.…”

Section: Introductionmentioning

confidence: 99%

An Abstraction-Based Framework for Neural Network Verification

Elboher

Gottschlich

Katz

2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Deep neural networks are increasingly being used as controllers for safety-critical systems. Because neural networks are opaque, certifying their correctness is a significant challenge. To address this issue, several neural network verification approaches have recently been proposed. However, these approaches afford limited scalability, and applying them to large networks can be challenging. In this paper, we propose a framework that can enhance neural network verification techniques by using over-approximation to reduce the size of the network—thus making it more amenable to verification. We perform the approximation such that if the property holds for the smaller (abstract) network, it holds for the original as well. The over-approximation may be too coarse, in which case the underlying verification tool might return a spurious counterexample. Under such conditions, we perform counterexample-guided refinement to adjust the approximation, and then repeat the process. Our approach is orthogonal to, and can be integrated with, many existing verification techniques. For evaluation purposes, we integrate it with the recently proposed Marabou framework, and observe a significant improvement in Marabou’s performance. Our experiments demonstrate the great potential of our approach for verifying larger neural networks.

show abstract

The three pillars of machine programming

Cited by 41 publications

References 73 publications

LEA: A Learned Encoding Advisor for Column Stores

LEA: A Learned Encoding Advisor for Column Stores

Learned garbage collection

An Abstraction-Based Framework for Neural Network Verification

Contact Info

Product

Resources

About