Original title: Transforming hierarchical images to program expressions using deep networks
Authors: Křen, Tomáš
Document type: Research reports
Year: 2018
Language: eng
Series: Technical report, volume: V-1263
Abstract: We present a technique describing how to effectively train a neural network given an image to produce a formal description of the given image. The basic motivation of the proposed technique is an intention to design a new tool for automatic program synthesis capable of transforming sensory data (in our case static image, but generally a phenotype) to a formal code expression (i.e. syntactic tree of a program), such that the code (from evolutionary perspective a genotype) evaluates to a value that is similar to the input data, ideally identical. Our approach is partially based on our technique for generating program expressions in the context of typed functional genetic programming. We present promising results evaluating a simple image description language achieved with a deep network combining convolution encoder of images and recurrent decoder for generating program expressions in the sequential prefix notation and propose possible future applications.
Keywords: automatic program synthesis; deep networks; image processing
Project no.: GA18-23827S (CEP)
Funding provider: GA ČR
Rights: This work is protected under the Copyright Act No. 121/2000 Coll.

Institution: Institute of Computer Science AS ČR (web)
Original record: http://hdl.handle.net/11104/0292265

Permalink: http://www.nusl.cz/ntk/nusl-391553


The record appears in these collections:
Research > Institutes ASCR > Institute of Computer Science
Reports > Research reports