第6回人工生命研究会

開催予定: 2023年2月21日

The 6th Workshop of Artificial Life Japan (JSAI SIG-ALIFE)

2023年2月21日(火)に人工生命研究会第6回ワークショップを開催いたします。
(02/12更新:スケジュールを公開しました)

We are happy to announce the Alife-Japan Workshop #6, on Tuesday, February 21, 2023 (Tue).
(Update 02/12: We have published the workshop’s schedule)

ワークショップ概要

今回のワークショップの目的は、広く人工生命に関連する研究についての発表と議論、フィードバックの機会を提供することです。とくに、今年7月に北海道で開催されるALIFE2023に投稿を予定されている研究についても投稿を歓迎いたします。ALIFE2023への投稿の前に、論文の質を高めるフィードバックの機会になることを期待しています。

また、Google BrainのBert Chan氏、Yujin Tang氏、Yingtao Tian氏のEvoJaxチュートリアルセッションを企画しています。EvoJaxとは、進化計算や人工生命シミュレーションにも使用されている、ハードウェア・アクセラレーション技術により高速化されたNeuroevolutionのツールキットです。

ワークショップはオンライン・オンサイト参加可能なハイブリッド方式で行います。オンサイトの開催地はつくばエクスプレス線つくば駅周辺のco-enスペースです。

重要な〆切

  • 題目・著者情報:2月03日(金) 2月10日(金)
  • 原稿提出:2月16日(木)
  • 参加登録(発表なし):2月20日(月)
  • ワークショップ: 2月21日(火)

〆切はすべて23:59日本時間(GMT+9)。参加登録はこちらのフォームへ

Covid対策のため、オンサイト参加者は先着順で30名に制限されます。

スケジュール

  • 09:00-10:20: 第1発表セッション(オンライン発表)
  • 10:30-12:10: 第2発表セッション
  • 14:00-15:10: 第3発表セッション
  • 15:30-18:00: EvoJAXチュートリアル
全スケジュール表(英語)
TimeAuthorsInstitutionTitle
9:00 – 9:05Claus AranhaUniversity of TsukubaOpening
9:05 – 9:30Caspar Luc, Witkowski OlafCross LabsCommunicating through a narrow channel: Perceiving microbial affect over a game of Go
9:30 – 9:55Luiz Fernando S. E. Santos, Claus Aranha, André Carlos P. L. F. de CarvalhoUniversity of São PauloMultiagent city expansion incorporating land use and transport
9:55 – 10:20LYU XiangThe University of TokyoPreventing Anesthetic Awareness: Should We Use Benzodiazepines or BIS Monitors?
10:20 – 10:30Short Break
10:30 – 10:55Takahide Yoshida, Takashi IkegamiThe University of TokyoImitation and Emulation in Humanoid Robot Alter3
10:55 – 11:20Fabio TanakaUniversity of TsukubaEvolving soft robots in neighboring environments
11:20 – 11:45西村春輝, 岡瑞起University of Tsukubaピクセル単位で年齢推定するニューラルセルオートマトンの考案
11:45 – 12:10岩橋七海、須田幹大、岡部純弥、岡瑞起University of Tsukuba社会的ネットワークの多様な相互作用を捉えるエージェントベースモデルの提案
12:10 – 13:55Lunch Break
14:00 – 14:25Hanna YoshidaThe University of TokyoA Review of the Development and Applications of Evolutionary Computation
14:25 – 14:50Tatsuya Sasaki, Satoshi Uchida, Isamu Okada, Hitoshi YamamotoKoriyama Women’s CollegeIntegrated indirect reciprocity and the evolution of cooperation
14:50 – 15:15Bayanbat Anar Erdene, Reiji Suzuki, and Takaya AritaNagoya UniversityThe effects of inter-player competition on the evolution of Theory of Mind in the cooperative card game Hanabi
15:15 – 15:30Short Break
15:30 – 17:00Bert Chan, Yujin Tang, Yingtao TianGoogle BrainEvoJAX: Hardware-Accelerated Evolution-Based Artificial Life
NOTE: This schedule is tentative, and still subject to change.

発表の詳細

発表15分+質疑10分を予定しています。発表は日本語でも英語でも可能です。

発表者は2月16日までに予稿または論文を提出してください。原稿は英語のみ、ALIFE2023学会の以下のフォーマットにしたがってください。(Full Paper/Extended Abstractどちらでも可)

https://sites.google.com/view/alife-2023/calls/call-for-papers-extended-abstracts

ワークショップに提出された発表や原稿に対してフィードバックのしくみを予定している。

EvoJAXチュートリアル

タイトル: “EvoJAX: Hardware-Accelerated Evolution-Based Artificial Life”

EvoJAXチュートリアル概要(英語)

Recently, hardware accelerators have played an important role in advancing the state-of-the-art for deep learning, enabling rapid training of neural networks and shorter research iteration cycles for their development. Much of this progress is restricted to systems that rely on gradient descent, a highly effective optimization method when we provide it with a well-defined objective function. But in areas such as artificial life, complex systems, computational biology, and even classical physics, fitness or objective functions are usually difficult to formulate if not totally non-existent. Much of the interesting behaviors we observe take place near chaotic states, where a system is constantly transitioning between order and disorder. It can be argued that intelligent life and even civilization are all complex systems operating at the edge of chaos. If we wish to study these systems, we need more efficient methods to simulate and find solutions in them.


Artificial life contains various approaches for modeling life-as-we-know-it and life-as-it-could-be, and evolutionary computation has made great progress in developing methods for evolving artificial life creatures and generating emergent structures and behaviors. Compared to gradient descent, evolution-based methods excel in escaping from local optima and finding novel solutions with great diversity. More importantly, it is often the case that complex systems are non-differentiable, rendering gradient-based methods infeasible, and even in cases where both types of methods can be used, evolution would yield better results.


However, the progress of hardware-accelerated computational methods for evolution has not kept pace with machine learning. Much of computational evolution and simulations are still conducted using CPUs, largely ignoring the recent breakthroughs in hardware accelerators such as GPUs/TPUs. Recent work started to demonstrate the effectiveness of GPUs for evolutionary algorithms and artificial life, but so far such demonstrations are tailored for specific implementations, limiting their applicability to other cases.


In this tutorial, we show our effort to enable greater access to hardware accelerators for artificial life researchers. Specifically, we describe EvoJAX, a scalable, general purpose toolkit we developed that is open source and available publicly. Building on the JAX library, our toolkit enables evolutionary algorithms to work with artificial life models running in parallel across multiple TPU/GPUs. EvoJAX achieves very high performance by implementing the algorithm, policy and task all in NumPy, which is compiled just-in-time to run on accelerators.


We will also demonstrate how one can use the EvoJAX system and extend EvoJAX for artificial life research. We showcase several extensible examples of EvoJAX for a wide range of tasks, including “Lenia” (virtual creatures evolve inside continuous cellular automata), “Ant Colony” (virtual ants form collective behavior by following pheromone), “Flocking” (a school of fish learn to move in a coherent fashion, much like Boids). Finally, we will show that EvoJAX can find solutions to most of these tasks within minutes on GPU/TPUs, compared to hours or days when using CPUs. We believe our toolkit can significantly shorten the experimental iteration cycle for researchers working with evolutionary computation.


All of our tutorial is accompanied by live running of code in the form of Python notebooks, to show how adapting EvoJAX for novel use cases is made straightforward. With Google Colab for running Python notebooks, no extra software setup is needed.

発表者バイオ(英語)

Yujin Tang is a Senior Research Engineer in Google Brain Tokyo. His research interests include neuroevolution algorithms, reinforcement learning, robotics and their applications. He obtained his B.S in Computer Science at Shanghai Jiao Tong University and M.S in Engineering Science at Waseda University.


Yingtao Tian is a Research Scientist in Google Brain Tokyo. His research interests lie in generative models and representation learning, and their applications in image generation, natural language processing, knowledge base modeling, social network modeling, bioinformatics. Recently he particularly focused on the interdisciplinary research of creativity / humanities research and machine intelligence such as using either non-differentiable evolution strategies or differentiable operations. Prior to that, he obtained his PhD in Computer Science at Stony Brook University.


Bert Chan is a Research Engineer in Google Brain Tokyo. His research interests include artificial life and evolutionary computation. He discovered a complex adaptive system called Lenia, for which he received the Outstanding Publication of 2019 award by the International Society for Artificial Life. Bert is also an external collaborator at Inria Bordeaux, and a visiting scholar at the Institute for Advanced Study, Princeton. He received his B.Sc. in Computer Science from the Chinese University of Hong Kong and M.Sc. in Cognitive Science from Lund University.

参考:

Outline

The main goal of this workshop is to provide an opportunity to discuss and provide feedback on recent works of the Artificial Life community in Japan. In particular, we also welcome work that you plan to submit to the ALIFE2023 conference in Hokkaido. We hope this will be a chance to provide feedback and increase the quality of your paper before submission!

We will also have a hands-on tutorial from Bert Chan, Yujin Tang and Yingtao Tian from Google Brain about EvoJax. EvoJax is a NeuroEvolution ToolKit improved by hardware acceleration technologies, that is also used for evolutionary computation and alife simulations.

The workshop will be hybrid, with both on-site and online participation possible. On-site participation will be at the co-en event space, next to the Tsukuba Station of the Tsukuba Express line.

Important Dates

  • Submission of Title/Author for presentations: February 03 (Fri) February 10th (Fri)
  • Submission of Paper/Abstract for presentation: February 16 (Thu)
  • Registration deadline for non-presenters: February 20 (Mon)
  • Workshop: February 21 (Tue)

All deadlines are 23:59 JST (GMT+9).

The number of on-site participants is limited to 30 people. It will be managed on a first-come, first-served basis.

Schedule

  • 9:00 to 10:20: First Presentation Session (Online Presentations)
  • 10:30 to 12:10: Second Presentation Session
  • 14:00 to 15:10: Third Presentation Session
  • 15:30 to 18:00: EvoJax Tutorial
Full Schedule Table
TimeAuthorsInstitutionTitle
9:00 – 9:05Claus AranhaUniversity of TsukubaOpening
9:05 – 9:30Caspar Luc, Witkowski OlafCross LabsCommunicating through a narrow channel: Perceiving microbial affect over a game of Go
9:30 – 9:55Luiz Fernando S. E. Santos, Claus Aranha, André Carlos P. L. F. de CarvalhoUniversity of São PauloMultiagent city expansion incorporating land use and transport
9:55 – 10:20LYU XiangThe University of TokyoPreventing Anesthetic Awareness: Should We Use Benzodiazepines or BIS Monitors?
10:20 – 10:30Short Break
10:30 – 10:55Takahide Yoshida, Takashi IkegamiThe University of TokyoImitation and Emulation in Humanoid Robot Alter3
10:55 – 11:20Fabio TanakaUniversity of TsukubaEvolving soft robots in neighboring environments
11:20 – 11:45西村春輝, 岡瑞起University of Tsukubaピクセル単位で年齢推定するニューラルセルオートマトンの考案
11:45 – 12:10岩橋七海、須田幹大、岡部純弥、岡瑞起University of Tsukuba社会的ネットワークの多様な相互作用を捉えるエージェントベースモデルの提案
12:10 – 13:55Lunch Break
14:00 – 14:25Hanna YoshidaThe University of TokyoA Review of the Development and Applications of Evolutionary Computation
14:25 – 14:50Tatsuya Sasaki, Satoshi Uchida, Isamu Okada, Hitoshi YamamotoKoriyama Women’s CollegeIntegrated indirect reciprocity and the evolution of cooperation
14:50 – 15:15Bayanbat Anar Erdene, Reiji Suzuki, and Takaya AritaNagoya UniversityThe effects of inter-player competition on the evolution of Theory of Mind in the cooperative card game Hanabi
15:15 – 15:30Short Break
15:30 – 17:00Bert Chan, Yujin Tang, Yingtao TianGoogle BrainEvoJAX: Hardware-Accelerated Evolution-Based Artificial Life
NOTE: This schedule is tentative, and still subject to change.

Presentation Details

Presentations are 15min, plus 10min QA. The presentation can be in English or Japanese.

Presenters are invited to sent a pre-print manuscript by February 16th. The manuscript must be in English, following the format of the ALIFE 2023 conference below (either Full paper or Extended Abstract).

https://sites.google.com/view/alife-2023/calls/call-for-papers-extended-abstracts

We are planning a system to provide constructive feedback for all submissions at the workshop.

EvoJAX Tutorial

Title: “EvoJAX: Hardware-Accelerated Evolution-Based Artificial Life”

Full summary of the EvoJAX Tutorial

Recently, hardware accelerators have played an important role in advancing the state-of-the-art for deep learning, enabling rapid training of neural networks and shorter research iteration cycles for their development. Much of this progress is restricted to systems that rely on gradient descent, a highly effective optimization method when we provide it with a well-defined objective function. But in areas such as artificial life, complex systems, computational biology, and even classical physics, fitness or objective functions are usually difficult to formulate if not totally non-existent. Much of the interesting behaviors we observe take place near chaotic states, where a system is constantly transitioning between order and disorder. It can be argued that intelligent life and even civilization are all complex systems operating at the edge of chaos. If we wish to study these systems, we need more efficient methods to simulate and find solutions in them.


Artificial life contains various approaches for modeling life-as-we-know-it and life-as-it-could-be, and evolutionary computation has made great progress in developing methods for evolving artificial life creatures and generating emergent structures and behaviors. Compared to gradient descent, evolution-based methods excel in escaping from local optima and finding novel solutions with great diversity. More importantly, it is often the case that complex systems are non-differentiable, rendering gradient-based methods infeasible, and even in cases where both types of methods can be used, evolution would yield better results.


However, the progress of hardware-accelerated computational methods for evolution has not kept pace with machine learning. Much of computational evolution and simulations are still conducted using CPUs, largely ignoring the recent breakthroughs in hardware accelerators such as GPUs/TPUs. Recent work started to demonstrate the effectiveness of GPUs for evolutionary algorithms and artificial life, but so far such demonstrations are tailored for specific implementations, limiting their applicability to other cases.


In this tutorial, we show our effort to enable greater access to hardware accelerators for artificial life researchers. Specifically, we describe EvoJAX, a scalable, general purpose toolkit we developed that is open source and available publicly. Building on the JAX library, our toolkit enables evolutionary algorithms to work with artificial life models running in parallel across multiple TPU/GPUs. EvoJAX achieves very high performance by implementing the algorithm, policy and task all in NumPy, which is compiled just-in-time to run on accelerators.


We will also demonstrate how one can use the EvoJAX system and extend EvoJAX for artificial life research. We showcase several extensible examples of EvoJAX for a wide range of tasks, including “Lenia” (virtual creatures evolve inside continuous cellular automata), “Ant Colony” (virtual ants form collective behavior by following pheromone), “Flocking” (a school of fish learn to move in a coherent fashion, much like Boids). Finally, we will show that EvoJAX can find solutions to most of these tasks within minutes on GPU/TPUs, compared to hours or days when using CPUs. We believe our toolkit can significantly shorten the experimental iteration cycle for researchers working with evolutionary computation.


All of our tutorial is accompanied by live running of code in the form of Python notebooks, to show how adapting EvoJAX for novel use cases is made straightforward. With Google Colab for running Python notebooks, no extra software setup is needed.

Presenter Bios

Yujin Tang is a Senior Research Engineer in Google Brain Tokyo. His research interests include neuroevolution algorithms, reinforcement learning, robotics and their applications. He obtained his B.S in Computer Science at Shanghai Jiao Tong University and M.S in Engineering Science at Waseda University.


Yingtao Tian is a Research Scientist in Google Brain Tokyo. His research interests lie in generative models and representation learning, and their applications in image generation, natural language processing, knowledge base modeling, social network modeling, bioinformatics. Recently he particularly focused on the interdisciplinary research of creativity / humanities research and machine intelligence such as using either non-differentiable evolution strategies or differentiable operations. Prior to that, he obtained his PhD in Computer Science at Stony Brook University.


Bert Chan is a Research Engineer in Google Brain Tokyo. His research interests include artificial life and evolutionary computation. He discovered a complex adaptive system called Lenia, for which he received the Outstanding Publication of 2019 award by the International Society for Artificial Life. Bert is also an external collaborator at Inria Bordeaux, and a visiting scholar at the Institute for Advanced Study, Princeton. He received his B.Sc. in Computer Science from the Chinese University of Hong Kong and M.Sc. in Cognitive Science from Lund University.

References:

イベント一覧