Figures and Tables from this paper
- figure 1
- table 1
- figure 2
- figure 3
- figure 4
- figure 6
Topics
Behaviour Trees (opens in a new tab)Reinforcement Learning (opens in a new tab)Behavioral Realism (opens in a new tab)Graphics (opens in a new tab)Artificial Intelligence (opens in a new tab)Q-learning (opens in a new tab)
49 Citations
- Ismael Sagredo-OlivenzaP. P. Gómez-MartínM. A. Gómez-MartínP. González-Calero
- 2019
Computer Science
IEEE Transactions on Games
TBTs are behavior trees (BTs) generated from traces obtained in a game through PbD that facilitate the use of BTs by game designers and promote their authoring control on game AI.
- 32
- Qi ZhangJian YaoQuanjun YinYabing Zha
- 2018
Computer Science
Applied Sciences
A novel idea of dynamic constraint based on frequent sub-trees mining, which can accelerate evolution by protecting preponderant behavior sub-Trees from undesired crossover, is proposed and introduced into the evolving BTs with hybrid constraints (EBT-HC).
- 18 [PDF]
- Chenjing ZhaoChuanshuai Deng Xiaodong Yi
- 2023
Computer Science
ICMLC
This paper presents intelligent generation methods that directly represent the policies generated by Q-learning and its derived algorithms in the form of BTs to enhance the interpretability of RL.
- 1
- Mart KartasevAron Granberg
- 2019
Computer Science, Art
This thesis investigates ways to extend the use of Reinforcement Learning (RL) to Behavior Trees (BTs) by using existing general-purpose RL methods within the framework of BTs.
- 8
- Highly Influenced
- PDF
- Lei LiLei WangYuanzhi LiJ. Sheng
- 2021
Computer Science, Engineering
ICAART
This work investigates a general and extendable model of mixed behavior tree (MDRL-BT) upon the option framework where the hierarchical architecture simultaneously involves different deep reinforcement learning nodes and normal BT nodes.
- 5
- PDF
- Rafael MarquesMarçal de Sousa
- 2021
Computer Science
A novel decanonicalization algorithm that converts a policy obtained from RL into a human legible Behavior Tree (BT), an architecture praised for its modularity and reactivity that can assist designers in testing and debugging and allow the users to better predict its behavior, thus cutting on development time and costs.
- PDF
- Qi ZhangKai XuPeng JiaoQuanjun Yin
- 2018
Computer Science
2018 IEEE 7th Data Driven Control and Learning…
Preliminary experiments show that the proposed modified evolving behavior trees approach to model agent behavior as a BT outperforms standard evolving behavior tree by achieving better final behavior performance with less learning episodes.
- 7
- Qi ZhangLin SunPeng JiaoQuanjun Yin
- 2017
Computer Science
2017 4th International Conference on Systems and…
Preliminary experiments in a predator-prey simulation scenario show that MAXQ-BT can facilitate behavior trees generation easily for CGF to achieve better behavior performance than handcrafted products.
- 11
- Highly Influenced
- Mart KartasevJustin SalerPetter Ögren
- 2023
Computer Science
2023 IEEE/RSJ International Conference on…
This paper proposes a way to set up the RL problems, such that they do not only achieve each immediate subgoal, but also avoid violating the identified ACCs.
- Mart KartasevJustin SalerPetter Ögren
- 2021
Computer Science
ArXiv
The key idea of this letter is to improve performance of backward chained BTs by using the conditions identified in the theoretical convergence proof to setup RL problems for individual controllers to avoid violating the identified ACCs.
- 1
...
...
27 References
- G. PugaM. A. Gómez-MartínBelén Díaz-AgudoP. González-Calero
- 2008
Computer Science
AIIDE
This paper focuses on dynamic retrieval of behaviours taking into account the world state and the underlying goals to select the most appropriate state machine to guide the NPC behaviour.
- 28
- PDF
- Du XiaoqinL. QinghuaHan Jianjun
- 2009
Computer Science
2009 IEEE International Conference on Automation…
The result shows that this method can satisfy the need for controlling NPC and has faster convergence speed than flat reinforcement learning.
- 6
- Chong-U LimRobin BaumgartenS. Colton
- 2010
Computer Science
EvoApplications
The use of behaviour trees are used to design and develop an AI-controlled player for the commercial real-time strategy game DEFCON which was able to outperform the game’s original AI-bot more than 50% of the time.
- 143
- PDF
- Luis PeñaSascha OssowskiJ. SánchezS. Lucas
- 2012
Computer Science
2012 IEEE Conference on Computational…
Two new variants of a hybrid algorithm, named WEREWoLF and WERESARSA, that combine evolutionary techniques with reinforcement learning are introduced that allow a group of different reinforcement learning controllers to be recombined in an iterative process that uses both evolution and learning.
- 19
- PDF
- David LlansóM. A. Gómez-MartínP. González-Calero
- 2009
Computer Science
AIIDE
An architecture for building the AI of an NPC that extends the component-based approach, which represents the functionality of an entity as a collection of functionality-specific components, and is able to identify faulty behaviour trees at design time.
- 7
- PDF
- A. BartoS. Mahadevan
- 2003
Computer Science
Discret. Event Dyn. Syst.
This work reviews several approaches to temporal abstraction and hierarchical organization that machine learning researchers have recently developed and discusses extensions of these ideas to concurrent activities, multiagent coordination, and hierarchical memory for addressing partial observability.
- 555
- A. BartoS. Mahadevan
- 2003
Computer Science
Discret. Event Dyn. Syst.
This work reviews several approaches to temporal abstraction and hierarchical organization that machine learning researchers have recently developed and discusses extensions of these ideas to concurrent activities, multiagent coordination, and hierarchical memory for addressing partial observability.
- 1,246
- PDF
- Thomas G. Dietterich
- 2000
Computer Science
J. Artif. Intell. Res.
The paper presents an online model-free learning algorithm, MAXQ-Q, and proves that it converges with probability 1 to a kind of locally-optimal policy known as a recursively optimal policy, even in the presence of the five kinds of state abstraction.
- 1,660
- PDF
- Diego Perez LiebanaMiguel NicolauM. O’NeillA. Brabazon
- 2011
Computer Science
EvoApplications
The results obtained reinforce the applicability of evolutionary programming systems to the development of artificial intelligence in games, and in dynamic systems in general, illustrating their viability as an alternative to more standard AI techniques.
- 92
- PDF
- Satinder SinghT. JaakkolaMichael I. Jordan
- 1994
Computer Science
NIPS
This paper presents a function approximator based on a simple extension to state aggregation (a commonly used form of compact representation), namely soft state aggregation, a theory of convergence for RL with arbitrary, but fixed, softstate aggregation, and a novel intuitive understanding of the effect of state aggregation on online RL.
- 385
- PDF
...
...
Related Papers
Showing 1 through 3 of 0 Related Papers