1-范文东+LLM+Agents+-+DA.pdf

1、Finding the Scaling Laws of AgentsWendong FanCore Member of CAMEL-AITech Lead of Eigent AICAMEL-AI.orgAgent from 1986 Agents are mindless processes Agent by itself can only do some simple things Joining these agents in societies leads to true intelligenceWhat magical trick makes us intelligent?The t

2、rick is that there is no trick.The power of intelligence stems from our vast diversity,not from any single,perfect principle.Marvin Minsky,The Society of Mind,p.308The agent-environment interaction in RLThe learner and decision-maker is called the AgentAgent interacts with an environment through act

3、ionsEnvironment presents new states and rewardsAgentsAgent in Reinforcement LearningGame AgentAutomated VehiclesGo GameAgentsAgent in Reinforcement LearningLilian Weng:https:/lilianweng.github.io/posts/2023-06-23-agent/Key differences:Language as InputLanguage as OutputState and Action are expressed

4、 as natural languageGeneralizability.Language Models as AgentsLanguage Models as AgentsKey Features:-Memory:Manages chat history and context window-Tools:Supports both internal and external function calls-Step Loop:Handle task require multiple request with one stepclass TaskSpecifyAgent(ChatAgent):A

5、n agent that specifies a given task prompt by prompting the user toprovide more details class TaskPlannerAgent(ChatAgent):An agent that helps divide a task into subtasks based on the inputtask prompt class TaskCreationAgent(ChatAgent):An agent that helps create new tasks based on the objectiveand la

6、st completed task.Compared to:obj:TaskPlannerAgent,its still a task planner,but it has more context informationlike last task and incomplete task list.class TaskPrioritizationAgent(ChatAgent):An agent that helps re-prioritize the task list andreturns numbered prioritized list.from camel.agents impor