Innateness, AlphaZero, and Artificial Intelligence

Abstract

The concept of innateness is rarely discussed in the context of artificial intelligence. When it is discussed, or hinted at, it is often the context of trying to reduce the amount of innate machinery in a given system. In this paper, I consider as a test case a recent series of papers by Silver et al (Silver et al., 2017a) on AlphaGo and its successors that have been presented as an argument that a "even in the most challenging of domains: it is possible to train to superhuman level, without human examples or guidance", "starting tabula rasa." I argue that these claims are overstated, for multiple reasons. I close by arguing that artificial intelligence needs greater attention to innateness, and I point to some proposals about what that innateness might look like.

References in corpus (3)

• Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
• Deep Learning: A Critical Appraisal
• Programmable Agents

Cited by in corpus (11)

• Neo: A Learned Query Optimizer
• Bao: Learning to Steer Query Optimizers
• Revisiting the poverty of the stimulus: hierarchical generalization without a hierarchical bias in recurrent neural networks
• Leveling the Playing Field -- Fairness in AI Versus Human Game Benchmarks
• A Roadmap to Domain Knowledge Integration in Machine Learning
• Transfer of Fully Convolutional Policy-Value Networks Between Games and Game Variants
• Deep learning for pedestrians: backpropagation in CNNs
• A Group-Theoretic Approach to Computational Abstraction: Symmetry-Driven Hierarchical Clustering
• Evolving Self-supervised Neural Networks: Autonomous Intelligence from Evolved Self-teaching
• Generalisation in Neural Networks Does not Require Feature Overlap
• Roadmap to a Roadmap: How Could We Tell When AGI is a 'Manhattan Project' Away?