Retaining Experience and Growing Solutions
May 06, 2015 ยท Declared Dead ยท ๐ arXiv.org
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Robyn Ffrancon
arXiv ID
1505.01474
Category
cs.NE: Neural & Evolutionary
Citations
0
Venue
arXiv.org
Last Checked
4 months ago
Abstract
Generally, when genetic programming (GP) is used for function synthesis any valuable experience gained by the system is lost from one problem to the next, even when the problems are closely related. With the aim of developing a system which retains beneficial experience from problem to problem, this paper introduces the novel Node-by-Node Growth Solver (NNGS) algorithm which features a component, called the controller, which can be adapted and improved for use across a set of related problems. NNGS grows a single solution tree from root to leaves. Using semantic backpropagation and acting locally on each node in turn, the algorithm employs the controller to assign subsequent child nodes until a fully formed solution is generated. The aim of this paper is to pave a path towards the use of a neural network as the controller component and also, separately, towards the use of meta-GP as a mechanism for improving the controller component. A proof-of-concept controller is discussed which demonstrates the success and potential of the NNGS algorithm. In this case, the controller constitutes a set of hand written rules which can be used to deterministically and greedily solve standard Boolean function synthesis benchmarks. Even before employing machine learning to improve the controller, the algorithm vastly outperforms other well known recent algorithms on run times, maintains comparable solution sizes, and has a 100% success rate on all Boolean function synthesis benchmarks tested so far.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
๐ Similar Papers
In the same crypt โ Neural & Evolutionary
๐ฎ
๐ฎ
The Ethereal
R.I.P.
๐ป
Ghosted
Deep Learning using Rectified Linear Units (ReLU)
R.I.P.
๐ป
Ghosted
Generative Adversarial Text to Image Synthesis
R.I.P.
๐ป
Ghosted
Regularized Evolution for Image Classifier Architecture Search
R.I.P.
๐ป
Ghosted
Temporal Ensembling for Semi-Supervised Learning
๐
๐
Old Age
Learning Structured Sparsity in Deep Neural Networks
Died the same way โ ๐ป Ghosted
R.I.P.
๐ป
Ghosted
Federated Learning: Strategies for Improving Communication Efficiency
R.I.P.
๐ป
Ghosted
In-Datacenter Performance Analysis of a Tensor Processing Unit
R.I.P.
๐ป
Ghosted
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning
R.I.P.
๐ป
Ghosted