Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Application of Quantum Annealing to Training of Deep Neural Networks #11

Open
minatoyuichiro opened this issue Jul 12, 2018 · 0 comments
Labels
Annealing Quantum Annealing MachineLearning 量子機械学習

Comments

@minatoyuichiro
Copy link

一言でいうと

従来のCD法などの古典近似計算を行わず、D-Waveのマシンから直接サンプリングを行い確率分布モデルのRBMを学習させた。

論文リンク

https://arxiv.org/abs/1510.06356

著者/所属機関

Steven H. Adachi1, Maxwell P. Henderson2
1 Lockheed Martin Information Systems & Global Solutions, Palo Alto, California, USA
(steven.h.adachi@lmco.com)
2 Lockheed Martin Information Systems & Global Solutions, King of Prussia, Pennsylvania, USA

投稿日付(yyyy/MM/dd)

2015/10/21

概要

D-Waveの実機を使用し、ディープラーニングの事前学習に使用される生成モデルの学習を行なった。これまではCD法などの近似サンプリングによって精度を落としていた計算に、直接サンプリングのアルゴリズムを適用することで、質の高い学習を行うことができた。

新規性・差分

近似計算では計算量も増え、モデルとの乖離もあった部分に、ダイレクトにD-Waveのアナログ性を用いた量子サンプリングを適用することで学習を大幅に効率化した。

手法

の期待値計算に直接量子サンプリングを適用し、N回の平均値だけで簡単に求まるモデルを適用した。

結果

古典計算の難しい部分を量子サンプリングで大幅に時間短縮した。

コメント

とてもいい論文。

@gyu-don gyu-don added Annealing Quantum Annealing MachineLearning 量子機械学習 labels Jul 15, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Annealing Quantum Annealing MachineLearning 量子機械学習
Projects
None yet
Development

No branches or pull requests

2 participants