Learning from Shader Program Traces

02/08/2021
by   Yuting Yang, et al.
0

Deep networks for image processing typically learn from RGB pixels. This paper proposes instead to learn from program traces, the intermediate values computed during program execution. We study this idea in the context of pixel shaders – programs that generate images, typically running in parallel (for each pixel) on GPU hardware. The intermediate values computed at each pixel during program execution form the input to the learned model. In a variety of applications, models learned from program traces outperform baseline models learned from RGB, even when augmented with hand-picked shader-specific features. We also investigate strategies for selecting a subset of trace features for learning; using just a small subset of the trace still outperforms the baselines.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 1

page 6

page 15

page 16

page 17

page 18

01/15/2020

Learning Concise Models from Long Execution Traces

Abstract models of system-level behaviour have applications in design ex...
07/30/2017

Learning to Infer Graphics Programs from Hand-Drawn Images

We introduce a model that learns to convert simple hand drawings into gr...
07/11/2019

Trace-Relating Compiler Correctness and Secure Compilation

Compiler correctness is, in its simplest form, defined as the inclusion ...
09/04/2021

Direct Construction of Program Alignment Automata for Equivalence Checking

The problem of checking whether two programs are semantically equivalent...
01/08/2020

Learning to Encode and Classify Test Executions

The challenge of automatically determining the correctness of test execu...
07/10/2018

Understanding Differences among Executions with Variational Traces

One of the main challenges of debugging is to understand why the program...
03/18/2018

Code Vectors: Understanding Programs Through Embedded Abstracted Symbolic Traces

With the rise of machine learning, there is a great deal of interest in ...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.