GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → dhruvramani → Neural Architecture Search With Rl

dhruvramani / Neural Architecture Search With Rl

Minimal Tensorflow implementation of the paper "Neural Architecture Search With Reinforcement Learning" presented at ICLR 2017

Programming Languages

python
139335 projects - #7 most used programming language

Labels

tensorflowneural-networkreinforcement-learningpaperresearchneural-architecture-search

Projects that are alternatives of or similar to Neural Architecture Search With Rl

GuidedLabelling
Exploiting Saliency for Object Segmentation from Image Level Labels, CVPR'17
Stars: ✭ 35 (-5.41%)
Mutual labels:  research, paper
Research Method
论文写作与资料分享
Stars: ✭ 436 (+1078.38%)
Mutual labels:  paper, research
Dreamerv2
Mastering Atari with Discrete World Models
Stars: ✭ 287 (+675.68%)
Mutual labels:  research, reinforcement-learning
cerberus research
Research tools for analysing Cerberus banking trojan.
Stars: ✭ 110 (+197.3%)
Mutual labels:  research, paper
Dnc Tensorflow
A TensorFlow implementation of DeepMind's Differential Neural Computers (DNC)
Stars: ✭ 587 (+1486.49%)
Mutual labels:  paper, research
best AI papers 2021
A curated list of the latest breakthroughs in AI (in 2021) by release date with a clear video explanation, link to a more in-depth article, and code.
Stars: ✭ 2,740 (+7305.41%)
Mutual labels:  research, paper
Learning Deep Learning
Paper reading notes on Deep Learning and Machine Learning
Stars: ✭ 388 (+948.65%)
Mutual labels:  paper, reinforcement-learning
Awesome Deeplearning Resources
Deep Learning and deep reinforcement learning research papers and some codes
Stars: ✭ 2,483 (+6610.81%)
Mutual labels:  paper, reinforcement-learning
Habitat Lab
A modular high-level library to train embodied AI agents across a variety of tasks, environments, and simulators.
Stars: ✭ 587 (+1486.49%)
Mutual labels:  research, reinforcement-learning
Qlib
Qlib is an AI-oriented quantitative investment platform, which aims to realize the potential, empower the research, and create the value of AI technologies in quantitative investment. With Qlib, you can easily try your ideas to create better Quant investment strategies. An increasing number of SOTA Quant research works/papers are released in Qlib.
Stars: ✭ 7,582 (+20391.89%)
Mutual labels:  research, paper
Neural-Architecture-Search
This repo is about NAS
Stars: ✭ 26 (-29.73%)
Mutual labels:  paper, neural-architecture-search
Gibsonenv
Gibson Environments: Real-World Perception for Embodied Agents
Stars: ✭ 666 (+1700%)
Mutual labels:  research, reinforcement-learning
Browser Sec Whitepaper
Cure53 Browser Security White Paper
Stars: ✭ 251 (+578.38%)
Mutual labels:  paper, research
adage
Data and code related to the paper "ADAGE-Based Integration of Publicly Available Pseudomonas aeruginosa..." Jie Tan, et al · mSystems · 2016
Stars: ✭ 61 (+64.86%)
Mutual labels:  research, paper
Research Paper Notes
Notes and Summaries on ML-related Research Papers (with optional implementations)
Stars: ✭ 218 (+489.19%)
Mutual labels:  paper, research
Lagom
lagom: A PyTorch infrastructure for rapid prototyping of reinforcement learning algorithms.
Stars: ✭ 364 (+883.78%)
Mutual labels:  research, reinforcement-learning
Epg
Code for the paper "Evolved Policy Gradients"
Stars: ✭ 204 (+451.35%)
Mutual labels:  paper, reinforcement-learning
Research In Production
A collection of research papers categorized by real-world systems that enact them
Stars: ✭ 205 (+454.05%)
Mutual labels:  paper, research
Holodeck
High Fidelity Simulator for Reinforcement Learning and Robotics Research.
Stars: ✭ 513 (+1286.49%)
Mutual labels:  research, reinforcement-learning
Dl Nlp Readings
My Reading Lists of Deep Learning and Natural Language Processing
Stars: ✭ 656 (+1672.97%)
Mutual labels:  paper, reinforcement-learning
View All Similar Projects ➔

Neural Architecture Search with Reinforcement Learning

model_lenet and model_dummy have different code. Check out model_lenet for better explanation.

Dataset

The following model was trained on the CIFAR-10 dataset. To get the data and data-prep. related filed contact me.

Working

This application is an implementation of Neural Architecture Search which uses a recurrent neural network to generate the hyperparameters. We use a softmax layer, and let the network "choose" between multiple choices which we provide (hard-coded), and construct a architecture which is trained, and the validation accuracy is calculated. The validation accuracy is used as the reward signal, and the goal is to maximize using it. For this, we use Gradient Ascent, and calculate the gradients using the REINFORCE algorithm.

I have tried out 2 implementations (in seperate folders). This is a minimal implementation of the algorithm and it's highly likely that it might not work.

Installation

This script was built and tested on python3, so make sure you use pip3! Install Tensorflow from the official website, that should install all the other dependencies too, hopefully.

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].