王昊 (2022-11-30 19:37):
#paper https://cis.temple.edu/tagit/presentations/A%20Path%20Towards%20Autonomous%20Machine%20Intelligence.pdf. A Path Towards Autonomous Machine Intelligence.  LeCun. A Path Towards Autonomous Machine Intelligence Version 0.9.2, 2022-06-27. 62. Yann LeCun指明下一代AI方向:自主机器智能。 LeCun在本文中提出了一套认知的架构,以及训练其中world model的方法。主要包括以下模块: (1)配置器(Configurator)模块负责执行控制(executive control):给定要执行的任务,可以通过调整这些模块的参数来预先配置感知模块(perception module)、世界模型(world model)、成本(cost)和当前任务的 actor。(2)感知模块(Perception module)接收来自传感器的信号并估计当前世界的状态,对于给定的任务,只有一小部分感知到的世界状态是相关和有用的。配置器模块启动感知系统,从感知中提取相关信息,完成手头的任务。(3)世界模型(World model)的作用是双重的:(1)估计感知未提供的关于世界状态的缺失信息;(2)预测合理的未来世界状态。(4)成本模块(Cost module)计算单个标量的输出,该输出预测智能体的不适(discomfort)程度。(5)Actor 模块计算动作序列的提议。(6)短期记忆模块(Short-term memory module)跟踪当前和预测的世界状态以及相关成本。
2022.
A Path Towards Autonomous Machine Intelligence
翻译
Abstract:
How could machines learn as efficiently as humans and animals? How could machines learn to reason and plan? How could machines learn representations of percepts and action plans at multiple levels of abstraction, enabling them to reason, predict, and plan at multiple time horizons? This position paper proposes an architecture and training paradigms with which to construct autonomous intelligent agents. It combines concepts such as configurable predictive world model, behavior driven through intrinsic motivation, and hierarchical joint embedding architectures trained with self-supervised learning.
翻译
回到顶部