cvpaper.challenge の Meta Study Group 発表スライド
cvpaper.challenge はコンピュータビジョン分野の今を映し、トレンドを創り出す挑戦です。論文サマリ?アイディア考案?議論?実装?論文投稿に取り組み、凡ゆる知識を共有します。2019の目標「トップ会議30+本投稿」「2回以上のトップ会議網羅的サーベイ」
http://xpaperchallenge.org/cv/
Self Introduction for people interested in me.孝好 飯塚
?
- The document discusses the career and accomplishments of Takayoshi Iitsuka, including his work developing compiler technologies for supercomputers at Hitachi and contributions to several national computing projects in Japan.
- After Hitachi shifted focus to storage research in 2001, Iitsuka led a project that improved the performance of NAS (Network Attached Storage) systems by 3 times through profiling and optimizations.
- Since retiring from Hitachi in 2015, Iitsuka has engaged in learning web technologies, deep learning, and achieving success with Montezuma's Revenge using reinforcement learning algorithms. He shares his work through blogs, GitHub repositories, and presentations.
This document contains summaries of various projects and analyses posted on GitHub by Takayoshi Iitsuka. It describes analyzing the intermediate layers of VAE models, approximating the distribution of MNIST data in high-dimensional spaces, using preemptible VMs on GCP, applying k-means clustering to MNIST, creating executable Python tutorial scripts, scraping templates, developing a "Pico-OS" for MicroPython, and speeding up recalculations in large Excel sheets. Links to the relevant GitHub repositories are provided.
cvpaper.challenge の Meta Study Group 発表スライド
cvpaper.challenge はコンピュータビジョン分野の今を映し、トレンドを創り出す挑戦です。論文サマリ?アイディア考案?議論?実装?論文投稿に取り組み、凡ゆる知識を共有します。2019の目標「トップ会議30+本投稿」「2回以上のトップ会議網羅的サーベイ」
http://xpaperchallenge.org/cv/
Self Introduction for people interested in me.孝好 飯塚
?
- The document discusses the career and accomplishments of Takayoshi Iitsuka, including his work developing compiler technologies for supercomputers at Hitachi and contributions to several national computing projects in Japan.
- After Hitachi shifted focus to storage research in 2001, Iitsuka led a project that improved the performance of NAS (Network Attached Storage) systems by 3 times through profiling and optimizations.
- Since retiring from Hitachi in 2015, Iitsuka has engaged in learning web technologies, deep learning, and achieving success with Montezuma's Revenge using reinforcement learning algorithms. He shares his work through blogs, GitHub repositories, and presentations.
This document contains summaries of various projects and analyses posted on GitHub by Takayoshi Iitsuka. It describes analyzing the intermediate layers of VAE models, approximating the distribution of MNIST data in high-dimensional spaces, using preemptible VMs on GCP, applying k-means clustering to MNIST, creating executable Python tutorial scripts, scraping templates, developing a "Pico-OS" for MicroPython, and speeding up recalculations in large Excel sheets. Links to the relevant GitHub repositories are provided.
This document presents mathematical formulas for calculating gradients and updates in reinforcement learning. It defines a formula for calculating the gradient of a value function with respect to its parameters, a formula for calculating the gradient of a policy based on the reward and value, and a formula for calculating the gradient of a parameter vector that is a weighted combination of its previous value and the policy gradient.
Safe and Efficient Off-Policy Reinforcement Learningmooopan
?
This document summarizes the Retrace(λ) reinforcement learning algorithm presented by Remi Munos, Thomas Stepleton, Anna Harutyunyan and Marc G. Bellemare. Retrace(λ) is an off-policy multi-step reinforcement learning algorithm that is safe (converges for any policy), efficient (makes best use of samples when policies are close), and has lower variance than importance sampling. Empirical results on Atari 2600 games show Retrace(λ) outperforms one-step Q-learning and existing multi-step methods.
Introduction of "TrailBlazer" algorithmKatsuki Ohto
?
論文「Blazing the trails before beating the path: Sample-efficient Monte-Carlo planning」紹介スライドです。NIPS2016読み会@PFN(2017/1/19) https://connpass.com/event/47580/ にて。
CluBERT: A Cluster-Based Approach for Learning Sense Distributions in Multipl...禎晃 山崎
?
CluBERT: A Cluster-Based Approach for Learning Sense Distributions in Multiple Languages
Word Sense Disambiguation, BERT, clustering
ということで読みました.
p. 7 は「solid は glass の上位語,glassware は glass の下位語」でした。。。