小柯机器人

美国纽约大学Brenden M. Lake和Marco Baroni通过元学习神经网络实现类人系统泛化。相关论文于2023年10月25日在线发表在《自然》杂志上。

研究人员成功地解决了Fodor和Pylyshyn的难题，提供了神经网络在优化组合技能时可以实现类人系统性的证据。为此，研究人员引入了元学习合成（MLC）方法，通过动态的合成任务流指导训练。为了比较人类和机器，研究人员使用指令学习范式进行了人类行为实验。

在考虑了七种不同的模型后，研究人员发现，与完全系统化但僵化的概率符号模型和完全灵活但非系统化的神经网络相比，只有MLC能够同时实现类人泛化所需的系统性和灵活性。在几个系统化泛化基准测试中，MLC还提高了机器学习系统的组成技能。这些研究结果表明，一个标准的神经网络架构在优化了其组成技能后，可以在正面比较中模仿人类的系统泛化能力。

据了解，人类语言和思维的力量源于系统的组合性：从已知成分中理解和产生新组合的代数能力。Fodor和Pylyshyn曾提出一个著名的观点，认为人工神经网络缺乏这种能力，因此不能作为思维的可行模型。此后几年，神经网络取得了长足的进步，但系统性难题依然存在。

附：英文原文

Title: Human-like systematic generalization through a meta-learning neural network

Author: Lake, Brenden M., Baroni, Marco

Issue&Volume: 2023-10-25

Abstract: The power of human language and thought arises from systematic compositionality—the algebraic ability to understand and produce novel combinations from known components. Fodor and Pylyshyn1 famously argued that artificial neural networks lack this capacity and are therefore not viable models of the mind. Neural networks have advanced considerably in the years since, yet the systematicity challenge persists. Here we successfully address Fodor and Pylyshyn’s challenge by providing evidence that neural networks can achieve human-like systematicity when optimized for their compositional skills. To do so, we introduce the meta-learning for compositionality (MLC) approach for guiding training through a dynamic stream of compositional tasks. To compare humans and machines, we conducted human behavioural experiments using an instruction learning paradigm. After considering seven different models, we found that, in contrast to perfectly systematic but rigid probabilistic symbolic models, and perfectly flexible but unsystematic neural networks, only MLC achieves both the systematicity and flexibility needed for human-like generalization. MLC also advances the compositional skills of machine learning systems in several systematic generalization benchmarks. Our results show how a standard neural network architecture, optimized for its compositional skills, can mimic human systematic generalization in a head-to-head comparison.

DOI: 10.1038/s41586-023-06668-3

Source: https://www.nature.com/articles/s41586-023-06668-3

Nature：《自然》，创刊于1869年。隶属于施普林格·自然出版集团，最新IF：69.504
官方网址：http://www.nature.com/
投稿链接：http://www.nature.com/authors/submit_manuscript.html