This document discusses self-supervised representation learning (SRL) for reinforcement learning tasks. SRL learns state representations by using prediction tasks as an auxiliary objective. The key ideas are: (1) SRL learns an encoder that maps observations to states using a prediction task like modeling future states or actions; (2) The learned state representations improve generalization and exploration in reinforcement learning algorithms; (3) Several SRL methods are discussed, including world models, inverse models, and causal infoGANs.
論文紹介 Anomaly Detection using One-Class Neural Networks (修正版Katsuki Ohto
?
This document discusses anomaly detection using one-class neural networks (OC-NN). It begins by introducing one-class support vector machines (OC-SVM) which learn a decision boundary to distinguish normal data points from anomalies using only normal data for training. The document then presents OC-NN as an alternative, where a neural network is trained to learn a low-dimensional representation of only normal data, and anomalies are detected as points with a large reconstruction error. It evaluates OC-NN on several datasets, finding it can achieve good performance compared to OC-SVM at detecting anomalies, as measured by the area under the ROC curve metric.
本スライドは、弊社の梅本により弊社内の技術勉強会で使用されたものです。
近年注目を集めるアーキテクチャーである「Transformer」の解説スライドとなっております。
"Arithmer Seminar" is weekly held, where professionals from within and outside our company give lectures on their respective expertise.
The slides are made by the lecturer from outside our company, and shared here with his/her permission.
Arithmer株式会社は東京大学大学院数理科学研究科発の数学の会社です。私達は現代数学を応用して、様々な分野のソリューションに、新しい高度AIシステムを導入しています。AIをいかに上手に使って仕事を効率化するか、そして人々の役に立つ結果を生み出すのか、それを考えるのが私たちの仕事です。
Arithmer began at the University of Tokyo Graduate School of Mathematical Sciences. Today, our research of modern mathematics and AI systems has the capability of providing solutions when dealing with tough complex issues. At Arithmer we believe it is our job to realize the functions of AI through improving work efficiency and producing more useful results for society.
This document discusses self-supervised representation learning (SRL) for reinforcement learning tasks. SRL learns state representations by using prediction tasks as an auxiliary objective. The key ideas are: (1) SRL learns an encoder that maps observations to states using a prediction task like modeling future states or actions; (2) The learned state representations improve generalization and exploration in reinforcement learning algorithms; (3) Several SRL methods are discussed, including world models, inverse models, and causal infoGANs.
論文紹介 Anomaly Detection using One-Class Neural Networks (修正版Katsuki Ohto
?
This document discusses anomaly detection using one-class neural networks (OC-NN). It begins by introducing one-class support vector machines (OC-SVM) which learn a decision boundary to distinguish normal data points from anomalies using only normal data for training. The document then presents OC-NN as an alternative, where a neural network is trained to learn a low-dimensional representation of only normal data, and anomalies are detected as points with a large reconstruction error. It evaluates OC-NN on several datasets, finding it can achieve good performance compared to OC-SVM at detecting anomalies, as measured by the area under the ROC curve metric.
本スライドは、弊社の梅本により弊社内の技術勉強会で使用されたものです。
近年注目を集めるアーキテクチャーである「Transformer」の解説スライドとなっております。
"Arithmer Seminar" is weekly held, where professionals from within and outside our company give lectures on their respective expertise.
The slides are made by the lecturer from outside our company, and shared here with his/her permission.
Arithmer株式会社は東京大学大学院数理科学研究科発の数学の会社です。私達は現代数学を応用して、様々な分野のソリューションに、新しい高度AIシステムを導入しています。AIをいかに上手に使って仕事を効率化するか、そして人々の役に立つ結果を生み出すのか、それを考えるのが私たちの仕事です。
Arithmer began at the University of Tokyo Graduate School of Mathematical Sciences. Today, our research of modern mathematics and AI systems has the capability of providing solutions when dealing with tough complex issues. At Arithmer we believe it is our job to realize the functions of AI through improving work efficiency and producing more useful results for society.
11. テキストから画像を生成するGAN
GAN-INT-CLS: Generative Adversarial Text to Image Synthesis, ICML 2016 (link)
GANをテキストからの画像生成に初めて利用。 64x64の画像生成に成功。
GAWWN: Learning What and Where to Draw, NIPS 2016 (link)
どの位置にどの物体があるかを BoundingBoxで指定することができた。
StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks, ICCV 2017 (link), C. Ledig, et.al.
2ステージ訓練により 256x256の圧倒的な高解像度を生成。もやもや画像が 2ステージ目でくっきり。
TAC-GAN: Text Conditioned Auxiliary Classifier Generative Adversarial Network, arXiv 2017 (link)
訓練アシストのための auxiliary classifierを使用。同じテキストから多様なタイプの画像が生成できる。
StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networks, arXiv 2017 (link), H. Zhang, et.al.
StackGAN-v2と呼ばれ、tree-likeネットワークを使用
AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networks, CVPR 2018 (link) Tao Xu, et.al.
Attentionドリブンな方法で細部を生成できるように
AttnGAN以降はend-to-end学習系の論文が出ている。
FusedGAN: Semi-supervised FusedGAN for Conditional Image Generation, arXiv 2018 (link)
2ステージ訓練を End-to-Endで学習できるよう1 stageにfuseした。
HDGAN: Photographic Text-to-Image Synthesis with a Hierarchically-nested Adversarial Network, arXiv 2018 (link)
hierarchical-nestedネットワーク構造で、高解像度画像を end-to-endで学習した。
12. LAPGAN(Laplacian Pyramid of Generative Adversarial Network)
Deep Generative Image Models using a ?Laplacian Pyramid of Adversarial Networks, NIPS(2015)
低解像度と高解像度の画像の差を学習し、低解像度画像をもとに高解像度の画像を生成する。各解像度のGeneratorを
学習し、段階的に高解像度の画像を生成。
AlignDRAW(Align Deep Recurrent Attention Writer)
Generating Images from Captions with Attention, ICLR(2016)
VAE(Variational Auto Encoder)はエンコーダの出力が正規分布の平均と共分散行列であり、潜在変数からのデータを再
構成する確率的変分推論アルゴリズムによって訓練する。そのVAEを拡張し、Attention機構を再帰的に組み込んだ
Deep Recurrent Attention Writer(DRAW)を追加したのがAlignDRAWである。AlignDRAWによりテキスト内の各単語
ごとに画像パッチを生成し、反復的に画像を生成した。
しかしこれらのモデルが生成する画像は、単語レベルの情報を欠いてしまっている。
高解像度画像生成の従来手法