The detailed results are described at GitHub (in English):
https://github.com/jkatsuta/exp-18-1q
(maddpg/experiments/my_notes/のexp1 ~ exp6)
立教大学のセミナー資料(前篇)です。
資料後篇:
/JunichiroKatsuta/ss-108099542
ブログ(動画あり):
https://recruit.gmo.jp/engineer/jisedai/blog/multi-agent-reinforcement-learning/
The detailed results are described at GitHub (in English):
https://github.com/jkatsuta/exp-18-1q
(maddpg/experiments/my_notes/のexp7 ~ exp11)
立教大学のセミナー資料(後篇)です。
資料前篇:
/JunichiroKatsuta/ss-108099238
ブログ(動画あり):https://recruit.gmo.jp/engineer/jisedai/blog/multi-agent-reinforcement-learning2/
This document provides biographical information about an individual named hiromu.yakura. It states that they were born in 1996 and are currently 19 years old. They have been interested in computer security since age 15 and have experience with the Linux operating system as well as programming Android applications. They have received several awards between 2012-2015 for achievements in computer security competitions and hacker conferences.
The detailed results are described at GitHub (in English):
https://github.com/jkatsuta/exp-18-1q
(maddpg/experiments/my_notes/のexp7 ~ exp11)
立教大学のセミナー資料(後篇)です。
資料前篇:
/JunichiroKatsuta/ss-108099238
ブログ(動画あり):https://recruit.gmo.jp/engineer/jisedai/blog/multi-agent-reinforcement-learning2/
This document provides biographical information about an individual named hiromu.yakura. It states that they were born in 1996 and are currently 19 years old. They have been interested in computer security since age 15 and have experience with the Linux operating system as well as programming Android applications. They have received several awards between 2012-2015 for achievements in computer security competitions and hacker conferences.
Trendmicro Security Award 2012 Final PresentationHiromu Yakura
?
The document describes a new malware detection system for Android that uses SEAndroid and Jubatus. It summarizes the increasing threat of Android malware and limitations of current security applications. The proposed system would use SEAndroid's mandatory access controls and logging combined with Jubatus' real-time machine learning to detect unauthorized commands and determine if applications are malware, defending against zero-day attacks. While it depends on device support for SEAndroid, which no current devices have, it could also use other Linux security modules and some devices support TOMOYO Linux as an alternative. The goal is to improve Android security.
This document does not contain any substantive information. It appears to be a blank document or one filled with repetitive text that does not form meaningful sentences or convey useful information.
颁罢贵とは、世界的に有名な旗取り合戦(Capture The Flag)のことで、セキュリティ技術を競うコンテストの総称です。出題ジャンルは、暗号、バイナリ、ネットワーク、Web、プログラミングなど多岐に渡り、クイズ形式の問題の謎を解いたり、実験ネットワーク内で疑似的な攻防戦を行ったりするものです。セキュリティだけでなくプログラミングに関する知見も問われ、攻撃技術、防御技術、解析技術、暗号の知見、ネットワーク技術など、広範な知識と経験が必要となっています。CTFは総合的な問題解決力を磨く最適な競技と言えるでしょう。
CTFには既に20年近い歴史があり、ラスベガスのDEFCONでCTFが開催されたことをきっかけに、今やヨーロッパやアジア、オセアニアや南米など、各国で頻繁に競技会が開催されています。国際大会も多く、元祖DEFCONを筆頭に、マレーシアHack in the BOXや、韓国CODEGATEなど、世界中からチームがCTFに参戦して熱戦を繰り広げています。
日本では2000年代の前半に「運動会」、それに続く「セキュリティスタジアム」が開催されましたが、現在に至るまで数年の空白期間があり、普及啓発や人材育成という面で世界に後れを取っているのが現状です。先日JNSA賞を受賞した日本チームsutegoma2の孤軍奮闘のおかげで昨年ようやくDEFCON CTFの世界予選を2位で突破しましたが、残念ながら本選ではふるわず、まだまだ世界とは大きな格差があると言わざるを得ません。
そんな中、今年SECCON CTF実行委員会を発足し、日本全体のセキュリティ技術の底上げと人材発掘をはかる目的で、国内でも本格的にCTF競技会を開催することとしました。
PFN summer intern 2015
Ayaka Kume
Replication work on "Robots that can adapt like animals"
Antonie Cully, Jeff Clune, Danesh Tarapore and Jean–Baptiste Mouret
Nature 521, 503–507 (29 May 2015)
ICRA 2018 (IEEE International Conference on Robotics and Automation; https://icra2018.org/ )の参加速報を書きました。
この資料には下記の項目が含まれています。
?ICRA 2018の概要
?ICRA 2018での動向や気付き
?ICRAの重要技術/重要論??
?AIST関連の論文
?今後の方針
?論文まとめ(100本あります)
Human-AI communication for human-human communication / CHAI Workshop @ IJCAI ...Hiromu Yakura
?
Human-AI communication for human-human communication: Applying interpretable unsupervised anomaly detection to executive coaching
In this paper, we discuss the potential of applying unsupervised anomaly detection in constructing AI-based interactive systems that deal with highly contextual situations, i.e., human-human communication, in collaboration with domain experts. We reached this approach of utilizing unsupervised anomaly detection through our experience of developing a computational support tool for executive coaching, which taught us the importance of providing interpretable results so that expert coaches can take both the results and contexts into account. The key idea behind this approach is to leave room for expert coaches to unleash their open-ended interpretations, rather than simplifying the nature of social interactions to well-defined problems that are tractable by conventional supervised algorithms. In addition, we found that this approach can be extended to nurturing novice coaches; by prompting them to interpret the results from the system, it can provide the coaches with educational opportunities. Although the applicability of this approach should be validated in other domains, we believe that the idea of leveraging unsupervised anomaly detection to construct AI-based interactive systems would shed light on another direction of human-AI communication.
This document summarizes research on generating adversarial examples to fool AI systems. It discusses 3 key papers:
1. Goodfellow et al. (2015) which introduced the concept of adversarial examples and showed they are common for deep learning models.
2. Carlini & Wagner (2018) which generated audio adversarial examples targeting speech recognition systems. They added imperceptible noise to audio to change the transcription.
3. Shafahi et al. (2019) which showed that adversarial examples may be inevitable for any classifier with enough capacity, including future more robust ones. Generating undetectable physical perturbations remains a challenge.
The document summarizes several academic papers and presentations by Hiromu Yakura from 2020 back to 2011 on topics related to music recommendation, malware analysis, behavioral analysis, generating bugs to fool classifiers, and whether adversarial examples are inevitable. The document lists Yakura's institutional affiliations which include University of Tsukuba and AIST in Japan.
The document discusses dependency injection in Python applications. It covers separating an application into domains, infrastructure, and presentation layers. It also discusses using dependency injection with abstract classes, type annotations, and value objects to decouple components and enable testing. Examples are provided of implementing dependency injection in Python.
Robust Audio Adversarial Example for a Physical AttackHiromu Yakura
?
This document summarizes and cites research on adversarial examples against speech recognition systems. It discusses papers that generated audio adversarial examples to target attacks on speech-to-text models, characterized temporal dependencies in audio adversarial examples, and developed approaches for creating targeted audio adversarial examples against black box speech recognition systems.
This document discusses the history and types of erasers. It describes how early erasers were made of natural gum before the modern plastic and rubber erasers were developed. The document also examines different kinds of erasers like knead erasers for removing marks lightly and ink erasers. It looks at new technologies involving erasers and electronic devices, and speculates on the future forms erasers may take.
現地時間3月3日から10日にかけて、世界中のテレコムが注目するテクノロジーカンファレンスである「Mobile World Conference 2025」がバルセロナで開催されました。特に競争の激しいヨーロッパのマーケットでは、各社が生き残りをかけたイノベーションをたくさん生み出しています。5G/6G、エッジクラウド、新しい音声技術など、多くのキーワードが注目されています。