Building Machines that Learn and Think for Themselves: Commentary on Lake et al., Behavioral and Brain Sciences, 2017
Building Machines that Learn and Think for Themselves: Commentary on Lake et al., Behavioral and Brain Sciences, 2017
复制标题
DOI：
--
复制
发表时间：
2017-11
期刊：
影响因子：
--
通讯作者：
M. Botvinick;D. Barrett;P. Battaglia;Nando de Freitas;D. Kumaran;Joel Z. Leibo;T. Lillicrap;Joseph Modayil;S. Mohamed;Neil C. Rabinowitz;Danilo Jimenez Rezende;Adam Santoro;T. Schaul;C. Summerfield;Greg Wayne;T. Weber;Daan Wierstra;S. Legg;D. Hassabis
中科院分区：
文献类型：
--
作者：
M. Botvinick;D. Barrett;P. Battaglia;Nando de Freitas;D. Kumaran;Joel Z. Leibo;T. Lillicrap;Joseph Modayil;S. Mohamed;Neil C. Rabinowitz;Danilo Jimenez Rezende;Adam Santoro;T. Schaul;C. Summerfield;Greg Wayne;T. Weber;Daan Wierstra;S. Legg;D. Hassabis
我们同意Lake及其同事关于构建类人智能的关键要素的清单，包括基于模型的推理是必不可少的。然而，我们倾向于一种以一个额外要素为中心的方法：自主性。特别是，我们的目标是代理，可以建立和利用自己的内部模型，以最小的人工工程。我们相信，当我们向现实世界的复杂性扩展时，以自主学习为中心的方法有最大的成功机会，解决现成的正式模型不可用的领域。在这里，我们调查了几个重要的例子，已经取得了进展，建立自主代理与人类一样的能力，并强调了一些突出的挑战。
We agree with Lake and colleagues on their list of key ingredients for building humanlike intelligence, including the idea that model-based reasoning is essential. However, we favor an approach that centers on one additional ingredient: autonomy. In particular, we aim toward agents that can both build and exploit their own internal models, with minimal human hand-engineering. We believe an approach centered on autonomous learning has the greatest chance of success as we scale toward real-world complexity, tackling domains for which ready-made formal models are not available. Here we survey several important examples of the progress that has been made toward building autonomous agents with humanlike abilities, and highlight some outstanding challenges.