NDC 2015 이은석 - pay-to-skip: 온라인 게임 속 로봇 경제와 내몰리는 인간Eunseok Yi
[NDC 2015 강연]
최근 인공지능(AI) 기술은 급격히 발전하고 있으며,
인간의 고유영역으로 생각됐던 분야들마저 더 효율좋은 기계가 점점 대체하고 있습니다.
머지 않은 미래에 로봇으로 인한 인간의 일자리 감소와, 자본주의 시스템의 부의 편중 문제는 훨씬 심각해질 것입니다.
한편, 인간사회의 축소판인 MMORPG에선 이런 일 역시 일찍 벌어지게 됩니다.
현실의 육체를 필요로 하지 않는 게임세계에서는 인간을 대신해 플레이하는 소위 '작업장'의 AI 봇(bot)들이 자칫 활개치기 쉬운데, 봇들은 돈을 내는 소수의 사용자를 위해 고용되므로 다수의 무료 유저들의 경쟁력을 떨어뜨려, 더욱 더 밀도 낮고 지루한 게임 경험을 하게 만들고, 결과적으로 Pay-to-Skip 게임이 돼버리게 합니다.
이런 현상의 메커니즘을 살펴보고, 문제를 완전히 해결하기는 어려워도 실마리를 찾아보고자 합니다.
NDC 2015 이은석 - pay-to-skip: 온라인 게임 속 로봇 경제와 내몰리는 인간Eunseok Yi
[NDC 2015 강연]
최근 인공지능(AI) 기술은 급격히 발전하고 있으며,
인간의 고유영역으로 생각됐던 분야들마저 더 효율좋은 기계가 점점 대체하고 있습니다.
머지 않은 미래에 로봇으로 인한 인간의 일자리 감소와, 자본주의 시스템의 부의 편중 문제는 훨씬 심각해질 것입니다.
한편, 인간사회의 축소판인 MMORPG에선 이런 일 역시 일찍 벌어지게 됩니다.
현실의 육체를 필요로 하지 않는 게임세계에서는 인간을 대신해 플레이하는 소위 '작업장'의 AI 봇(bot)들이 자칫 활개치기 쉬운데, 봇들은 돈을 내는 소수의 사용자를 위해 고용되므로 다수의 무료 유저들의 경쟁력을 떨어뜨려, 더욱 더 밀도 낮고 지루한 게임 경험을 하게 만들고, 결과적으로 Pay-to-Skip 게임이 돼버리게 합니다.
이런 현상의 메커니즘을 살펴보고, 문제를 완전히 해결하기는 어려워도 실마리를 찾아보고자 합니다.
The slides of Artificial Intelligence and Entertainment Science (AIES) Workshop 2021 Keynote lecture
https://aies.info/program/
Empathic Entertainment in Digital Game
A digital game give a unique experience to a user. AI system in Digital game consists of three kinds of AI such as Meta-AI, Character AI, and Spatial AI. Game experience is formed by them. Meta-AI keeps watching a status of game and controlling characters, objects, terrain, weather and so on dynamically to make many dramatic and empathic situations in a game for users. Character AI is a brain of an autonomous game character to make a decision by itself, but sometimes it acts to achieve a goal issued from Meta-AI. Spatial AI analyses a terrain and abstracts its features to communicate them to Meta-AI and Character-AI. They can make their intelligent decisions by using specific terrain and environment features. The AI system is called MCS-AI dynamic cooperative model (Meta-AI, Character AI, and Spatial AI dynamic cooperative model). In the lecture, I will explain the system by showing some cases of published digital games.
37. 3 第二次AIブーム(1980年代)
IF (A) then B
IF (C) then D
IF (E) then F
IF (G) then H
IF ( I ) then J
シンボルによる人工知能
(記号主義)
ニューラルネットによる人工知能
(コネクショニズム)
ルールベース
新しい学習法=
逆伝搬法
115. Two Agent Cooperation by DeepMind
Deep Mind: Capture the Flag: the emergence of complex cooperative agents
https://deepmind.com/blog/article/capture-the-flag-science
116. Deep Mind: Capture the flag
Deep Mind: Capture the Flag: the emergence of complex cooperative agents
https://deepmind.com/blog/article/capture-the-flag-science
117. Deep Q-Learning
• https://becominghuman.ai/lets-build-an-atari-ai-part-1-dqn-
df57e8ff3b26
Volodymyr Mnih, Koray Kavukcuoglu, David Silver, Alex Graves,
Ioannis Antonoglou, Daan Wierstra, Martin Riedmiller (DeepMind Technologies)
Playing Atari with Deep Reinforcement Learning
http://www.cs.toronto.edu/~vmnih/docs/dqn.pdf
画面を入力
操作はあらかじめ教える
スコアによる強化学習
https://becominghuman.ai/lets-build-an-atari-ai-part-1-dqn-df57e8ff3b26
118. • Pπ ロールアウトポリシー(ロールアウトで討つ手を決める。
Pπ(a|s) sという状態でaを討つ確率)
• Pσ Supervised Learning Network プロの討つ手からその
手を討つ確率を決める。Pσ(a|s)sという状態でaを討つ確
率。
• Pρ 強化学習ネットワーク。Pρ(学習済み)に初期化。
• Vθ(s’) 局面の状態 S’ を見たときに、勝敗の確率を予測
する関数。つまり、勝つか、負けるかを返します。
Mastering the game of Go with deep neural networks and tree search
http://www.nature.com/nature/journal/v529/n7587/full/nature16961.html
https://deepmind.com/research/alphago/
124. F-formation (Kendon, 1984)
• 人と人が向い合うときに、形成する立ち位置。
Paul Marshall,Yvonne Rogers,Nadia Pantidi
Using F-formations to analyse spatial patterns of interaction in physical environments
http://mcs.open.ac.uk/pervasive/pdfs/MarshallCSCW2011.pdf
156. Enhancing Game Experiences with Character AI
Andrew Moran, Jordan Carlton(Magic Leap, Magic Leap/Weta Workshop)
https://gdcvault.com/play/1025829/Magic-Leap-Enhancing-Game-Experiences
157. Enhancing Game Experiences with Character AI
Andrew Moran, Jordan Carlton(Magic Leap, Magic Leap/Weta Workshop)
https://gdcvault.com/play/1025829/Magic-Leap-Enhancing-Game-Experiences
158. Kaolin: A PyTorch Library for Accelerating 3D Deep Learning Research
https://arxiv.org/abs/1911.05063
159. Kaolin: A PyTorch Library for Accelerating 3D Deep Learning Research
https://arxiv.org/abs/1911.05063
166. 大型ゲームエンジン一覧(他にもたくさん)
タイトル ゲームエンジン名 会社
Far Cry 3,4 DUNIA ENGINE 2 Ubisoft Montreal
THE DIVISION snow drop engine Ubisoft (massive)
Assassin’s creed: syndicate AnvilNext 2.0 game engine Ubisoft Montreal
For Honor AnvilNext 2.0 game engine Ubisoft Montreal
Rise of Tomb Raider Foundation engine Crystal Dynamics
The Witcher 3 RED ENGINE CD PROJEKT
Dragon Age : Inquisition frostbite engine EA DICE
ゲームエンジン名 会社
汎用型 Unity3D Unity Technologies (デンマーク)
汎用型 UNREAL ENGINE 4 Epic Games (米)
汎用型 CryEngine CryTech (独)
汎用型 Lumberyard Amazon
汎用型 Stingray Autodesk
198. Enhancing Game Experiences with Character AI
Andrew Moran, Jordan Carlton(Magic Leap, Magic Leap/Weta Workshop)
https://gdcvault.com/play/1025829/Magic-Leap-Enhancing-Game-Experiences
199. Enhancing Game Experiences with Character AI
Andrew Moran, Jordan Carlton(Magic Leap, Magic Leap/Weta Workshop)
https://gdcvault.com/play/1025829/Magic-Leap-Enhancing-Game-Experiences
200. Kaolin: A PyTorch Library for Accelerating 3D Deep Learning Research
https://arxiv.org/abs/1911.05063
201. Kaolin: A PyTorch Library for Accelerating 3D Deep Learning Research
https://arxiv.org/abs/1911.05063
216. Paul van Grinsven; Player Traversal Mechanics in the Vast World of Horizon Zero Dawn (GDC2017)
https://www.guerrilla-games.com/read/player-traversal-mechanics-in-the-vast-world-of-horizon-zero-dawn
225. メタAI Left 4 Dead の事例
Michael Booth, "The AI Systems of Left 4 Dead," Artificial Intelligence and
Interactive Digital Entertainment Conference at Stanford.
http://www.valvesoftware.com/publications.html
今回は Left 4 Dead の事例を見てみる。
226. メタAI(=AI Director)によるユーザーのリラックス度に応じた敵出現度
ユーザーの緊張度
実際の敵出現数
計算によって
求められた
理想的な敵出現数
Build Up …プレイヤーの緊張度が目標値を超えるまで
敵を出現させ続ける。
Sustain Peak … 緊張度のピークを3-5秒維持するために、
敵の数を維持する。
Peak Fade … 敵の数を最小限へ減少していく。
Relax … プレイヤーたちが安全な領域へ行くまで、30-45秒間、
敵の出現を最小限に維持する。
Michael Booth, "The AI Systems of Left 4 Dead," Artificial Intelligence and Interactive Digital Entertainment Conference at Stanford.
http://www.valvesoftware.com/publications.html
より具体的なアルゴリズム
227. 安全な領域までの道のり(Flow Distance)
メタAIはプレイヤー群の経路を
トレースし予測する。
- どこへ来るか
- どこが背面になるか
- どこに向かうか
Michael Booth, "The AI Systems of Left 4 Dead," Artificial Intelligence and Interactive Digital Entertainment Conference at Stanford.
http://www.valvesoftware.com/publications.html
229. 敵出現領域
背後 前方
Michael Booth, "The AI Systems of Left 4 Dead," Artificial Intelligence and Interactive Digital Entertainment Conference at Stanford.
http://www.valvesoftware.com/publications.html
前方と背後のプレイヤー群から見えてない部屋に、
モンスターを発生させる。
230. Procedural Generation in WarFrame
• Warframe ではダンジョンが自動生成される。
Daniel Brewer, AI Postmortems: Assassin's Creed III, XCOM: Enemy Unknown, and Warframe (GDC2015)
http://www.gdcvault.com/play/1018223/AI-Postmortems-Assassin-s-Creed
231. Black Combination in WarFrame
• ブロックを組み合わる
• 完全に零からの生成
ではない。
このような生成のことを
Semi-procedural と言う。
Daniel Brewer, AI Postmortems: Assassin's Creed III, XCOM: Enemy Unknown, and Warframe (GDC2015)
http://www.gdcvault.com/play/1018223/AI-Postmortems-Assassin-s-Creed
234. スタートポイント、出口、目的地の
自動生成
Daniel Brewer, AI Postmortems: Assassin's Creed III, XCOM: Enemy Unknown, and Warframe (GDC2015)
http://www.gdcvault.com/play/1018223/AI-Postmortems-Assassin-s-Creed
235. ヒートマップ(影響マップ)を用いて
ゲーム中にプレイヤーの周囲を自動解析
Daniel Brewer, AI Postmortems: Assassin's Creed III, XCOM: Enemy Unknown, and Warframe (GDC2015)
http://www.gdcvault.com/play/1018223/AI-Postmortems-Assassin-s-Creed
ヒートマップ(影響マップ)とは、対象(ここではプレイヤー)を中心に、位置に温度(影響度)を
与える方法です。距離に応じて減衰します。また時間が経つと、周囲に熱が拡散します。
238. アクティブ・エリアセット(Active Are Set)
Daniel Brewer, AI Postmortems: Assassin's Creed III, XCOM: Enemy Unknown, and Warframe (GDC2015)
http://www.gdcvault.com/play/1018223/AI-Postmortems-Assassin-s-Creed
アクティブ・エリアセットは、プレイヤーの周囲の領域で、
リアルタイムにメタAIがゲームを調整する領域
240. メタAI (AI Director,)による
動的ペース調整
Daniel Brewer, AI Postmortems: Assassin's Creed III, XCOM: Enemy Unknown, and Warframe (GDC2015)
http://www.gdcvault.com/play/1018223/AI-Postmortems-Assassin-s-Creed
241. メタAI(自動適応ペーシング)
メタAI (AI Director,)による
動的ペース調整
Daniel Brewer, AI Postmortems: Assassin's Creed III, XCOM: Enemy Unknown, and Warframe (GDC2015)
http://www.gdcvault.com/play/1018223/AI-Postmortems-Assassin-s-Creed
242. メタAIによる出会うモンスターの数の大域調整
Daniel Brewer, AI Postmortems: Assassin's Creed III, XCOM: Enemy Unknown, and Warframe (GDC2015)
http://www.gdcvault.com/play/1018223/AI-Postmortems-Assassin-s-Creed
プレイヤーのスタート地点から出口までの道のりで、
コンスタントにモンスターと出会うようにする。
243. FarCry 4 の事例
Julien Varnier, Far Cry's AI: A Manifesto for Systemic Gameplay
http://archives.nucl.ai/recording/far-crys-ai-a-manifesto-for-systemic-gameplay/
244. FarCry 4 の事例
Julien Varnier, Far Cry's AI: A Manifesto for Systemic Gameplay
http://archives.nucl.ai/recording/far-crys-ai-a-manifesto-for-systemic-gameplay/
245. FarCry 4 の事例
Julien Varnier, Far Cry's AI: A Manifesto for Systemic Gameplay
http://archives.nucl.ai/recording/far-crys-ai-a-manifesto-for-systemic-gameplay/
246. FarCry 4 の事例
Julien Varnier, Far Cry's AI: A Manifesto for Systemic Gameplay
http://archives.nucl.ai/recording/far-crys-ai-a-manifesto-for-systemic-gameplay/
264. メタAI Left 4 Dead の事例
Michael Booth, "The AI Systems of Left 4 Dead," Artificial Intelligence and
Interactive Digital Entertainment Conference at Stanford.
http://www.valvesoftware.com/publications.html
今回は Left 4 Dead の事例を見てみる。
265. メタAI(=AI Director)によるユーザーのリラックス度に応じた敵出現度
ユーザーの緊張度
実際の敵出現数
計算によって
求められた
理想的な敵出現数
Build Up …プレイヤーの緊張度が目標値を超えるまで
敵を出現させ続ける。
Sustain Peak … 緊張度のピークを3-5秒維持するために、
敵の数を維持する。
Peak Fade … 敵の数を最小限へ減少していく。
Relax … プレイヤーたちが安全な領域へ行くまで、30-45秒間、
敵の出現を最小限に維持する。
Michael Booth, "The AI Systems of Left 4 Dead," Artificial Intelligence and Interactive Digital Entertainment Conference at Stanford.
http://www.valvesoftware.com/publications.html
より具体的なアルゴリズム