ToyArchitecture: Unsupervised Learning of Interpretable Models of the World

Abstract

Research in Artificial Intelligence (AI) has focused mostly on two extremes: either on small improvements in narrow AI domains, or on universal theoretical frameworks which are usually uncomputable, incompatible with theories of biological intelligence, or lack practical implementations. The goal of this work is to combine the main advantages of the two: to follow a big picture view, while providing a particular theory and its implementation. In contrast with purely theoretical approaches, the resulting architecture should be usable in realistic settings, but also form the core of a framework containing all the basic mechanisms, into which it should be easier to integrate additional required functionality. In this paper, we present a novel, purposely simple, and interpretable hierarchical architecture which combines multiple different mechanisms into one system: unsupervised learning of a model of the world, learning the influence of one's own actions on the world, model-based reinforcement learning, hierarchical planning and plan execution, and symbolic/sub-symbolic integration in general. The learned model is stored in the form of hierarchical representations with the following properties: 1) they are increasingly more abstract, but can retain details when needed, and 2) they are easy to manipulate in their local and symbolic-like form, thus also allowing one to observe the learning process at each level of abstraction. On all levels of the system, the representation of the data can be interpreted in both a symbolic and a sub-symbolic manner. This enables the architecture to learn efficiently using sub-symbolic methods and to employ symbolic inference.

Revision: changed the pdftitle

References in corpus (36)

• Intriguing properties of neural networks
• Deep Residual Learning for Image Recognition
• Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation
• Opening the Black Box of Deep Neural Networks via Information
• A simple neural network module for relational reasoning
• Why Neurons Have Thousands of Synapses, A Theory of Sequence Memory in Neocortex
• One-shot Learning with Memory-Augmented Neural Networks
• Neural-Symbolic Learning and Reasoning: A Survey and Interpretation
• Measuring the tendency of CNNs to Learn Surface Statistical Regularities
• Model-Free Episodic Control
• Unsupervised Predictive Memory in a Goal-Directed Agent
• DARLA: Improving Zero-Shot Transfer in Reinforcement Learning
• Early Visual Concept Learning with Unsupervised Deep Learning
• The Consciousness Prior
• Dynamic Routing Between Capsules
• Random feedback weights support learning in deep neural networks
• A Review of 40 Years of Cognitive Architecture Research: Core Cognitive Abilities and Practical Applications
• Depth-Gated LSTM
• Disentangling the independently controllable factors of variation by interacting with the world
• An Overview of Hierarchical Task Network Planning
• Cortical microcircuits as gated-recurrent neural networks
• Deep Predictive Learning: A Comprehensive Model of Three Visual Streams
• Towards Quantifying Complexity with Quantum Mechanics
• Sparsey: Event Recognition via Deep Hierarchical Spare Distributed Codes
• Hierarchical Reinforcement Learning with the MAXQ Value Function Decomposition
• Bounded Recursive Self-Improvement
• Unsupervised Learning from Continuous Video in a Scalable Predictive Recurrent Network
• The AGI Containment Problem
• CortexNet: a Generic Network Family for Robust Visual Temporal Representations
• A Framework for Searching for General Artificial Intelligence
• POWERPLAY: Training an Increasingly General Problem Solver by Continually Searching for the Simplest Still Unsolvable Problem
• A Neurally-Inspired Hierarchical Prediction Network for Spatiotemporal Sequence Learning and Prediction
• Fundamental principles of cortical computation: unsupervised learning with prediction, compression and feedback
• Feynman Machine: The Universal Dynamical Systems Computer
• Artificial general intelligence through recursive data compression and grounded reasoning: a position paper
• Hierarchical compositional feature learning