深度强化学习可以在共同的资源问题中促进可持续的人类行为
Raphael Koster1, Miruna Pîslar2, Andrea Tacchetti3
1Google DeepMind, London, UK. rkoster@google.com.
Nature communications
|March 23, 2025
概括
深度强化学习设计了一个社会规划器,用于在信任游戏中可持续地分享资源. 这种人工智能机制通过适应慷慨和惩罚叛逃者,促进了更公平,更大的集体回报.
科学领域:
- 行为经济学是一种行为经济学.
- 人工智能的人工智能
- 游戏理论 游戏理论
背景情况:
- 社会困境,如资源分配,往往涉及个人利益和集体福祉之间的冲突.
- 有效的资源分配机制对于维持共享资源和促进合作至关重要.
- 以前的机制往往无法适应动态的玩家行为和资源的可用性.
研究的目的:
- 设计一个基于人工智能 (AI) 的社会规划器,使用深度强化学习 (RL) 来促进多人信任游戏中的可持续贡献.
- 创建一个模拟经济的人类类玩家测试和完善RL机制.
- 为了最大限度地提高所有参与者的总回报,同时促进公平与合作.
主要方法:
- 训练有素的神经网络模拟人类玩家行为在一个代的多人信任游戏.
- 使用深度强化学习 (RL) 训练一个社会规划机制.
- 制定了一项再分配政策,将慷慨与资源的可用性联系在一起,并对叛逃者实施临时制裁.
主要成果:
- 与基线机制相比,RL设计的机制显著增加了总回报.
- 该机制实现了在参与者之间更平等地分配余.
- 人工智能的政策,当翻译成一个可解释的机制时,被人类参与者更好地接受.
结论:
- 深度强化学习可以有效地设计适应社会困境的适应机制,促进可持续的合作.
- 人工智能驱动的资源分配可以在共享资源管理中提高效率和提高公平性.
- 可解释的人工智能衍生政策提高了玩家接受和参与合作系统.
相关概念视频
01:24Generalization, Discrimination, and Extinction
375
Generalization, discrimination, and extinction are key concepts in operant conditioning that influence how behaviors are learned and maintained.
Generalization occurs when a behavior reinforced in one context is performed in similar situations. For instance, a student who studies diligently for calculus and receives excellent grades might apply the same study habits to psychology and history, expecting similar results. Generalization shows how learning in one setting can influence behavior in...
Generalization occurs when a behavior reinforced in one context is performed in similar situations. For instance, a student who studies diligently for calculus and receives excellent grades might apply the same study habits to psychology and history, expecting similar results. Generalization shows how learning in one setting can influence behavior in...
375
01:23Primary and Secondary Reinforcers
153
In psychology, reinforcement is a key concept in behavior modification. B.F. Skinner demonstrated this with his experiments involving rats in what is known as a Skinner box. The rats learned to press a lever to receive food, a primary reinforcer that fulfilled their innate need for nourishment.
Effective reinforcers for humans vary depending on the individual and the context. Primary reinforcers, such as food, water, sleep, shelter, and pleasure, have inherent value and satisfy basic biological...
Effective reinforcers for humans vary depending on the individual and the context. Primary reinforcers, such as food, water, sleep, shelter, and pleasure, have inherent value and satisfy basic biological...
153
01:23Reinforcement
169
Positive and negative reinforcement are key concepts in operant conditioning, a learning process where the consequences of a behavior affect the likelihood of that behavior being repeated.
Positive reinforcement occurs when a behavior is followed by the presentation of a rewarding stimulus, increasing the frequency of that behavior. For example:
Positive reinforcement occurs when a behavior is followed by the presentation of a rewarding stimulus, increasing the frequency of that behavior. For example:
169
01:05Instinctive Drift
169
Instinctive drift refers to the tendency of animals to revert to their innate behaviors despite repeated reinforcement. Breland and Breland demonstrated this concept in an experiment with a raccoon. The raccoon was trained to pick up two coins and place them in a container in exchange for food. Initially, the raccoon learned to associate the coins with food, making them a conditioned stimulus or a substitute for food. However, over time, the raccoon became less willing to put the coins into the...
169
01:08Timing and Consequences on Behavior
66
In operant conditioning, the timing of reinforcement is crucial. For animals like rats and cats, immediate reinforcement (within a few seconds) is much more effective than delayed reinforcement. For example, a food reward for a rat needs to follow within 30 seconds of pressing a bar to be effective.
Humans, however, can respond to delayed reinforcers. We often make decisions between immediate small rewards and delayed larger rewards. This ability to delay gratification is a significant...
Humans, however, can respond to delayed reinforcers. We often make decisions between immediate small rewards and delayed larger rewards. This ability to delay gratification is a significant...
66
01:24Operant Conditioning Intervention
32
Operant conditioning serves as a foundational principle in therapeutic interventions aimed at modifying maladaptive behaviors. Central to this approach is the notion that behaviors, both adaptive and maladaptive, are learned through reinforcement. By analyzing the environmental factors that reinforce problematic behaviors, clinicians can design interventions to weaken these reinforcements and replace maladaptive behaviors with healthier alternatives.
In operant conditioning, behaviors that are...
In operant conditioning, behaviors that are...
32