Building Machines that Learn and Think Like People

By Josh Tenenbaum. 2018

2,218 citations recorded by the source corpus.

Recent successes in artificial intelligence and machine learning have been largely driven by methods for sophisticated pattern recognition, including deep neural networks and other data-intensive methods. But human intelligence is more than just pattern recognition. And no machine system yet built has anything like the flexible, general-purpose commonsense grasp of the world that we can see in even a one-year-old human infant. I will consider how we might capture the basic learning and thinking abilities humans possess from early childhood, as one route to building more human-like forms of machine learning and thinking. At the heart of human common sense is our ability to model the physical and social environment around us: to explain and understand what we see, to imagine things we could see but haven't yet, to solve problems and plan actions to make these things real, and to build new models as we learn more about the world. I will focus on our recent work reverse-engineering these capacities using methods from probabilistic programming, program induction and program synthesis, which together with deep learning methods and video game simulation engines, provide a toolkit for the joint enterprise of modeling human intelligence and making AI systems smarter in more human-like ways.

Subfields: Artificial Intelligence, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Evaluation & Benchmarks

Citations in the corpus

How Can We Accelerate Progress Towards Human-like Linguistic Generalization? (2020, background): “Humans’ rapid and consistent generalization abilities rely on powerful inductive biases, which likely arise from a combination of innate building blocks and experience with diverse learning problems (Lake et al., 2017).”
Learning to learn by gradient descent by gradient descent (2016, background): “More recently, Lake et al. [2016] have argued forcefully for its importance as a building block in artiﬁcial intelligence.”
What can the brain teach us about building artificial intelligence? (2016, background): “In The Society of Mind, Minsky (1986) argued that the human brain is more similar to a complex society of diverse neural networks, than to a large, single one.”
Programmable Agents (2017, background): “Compositionality has become a topic of great interest in machine learning, robotics and cognitive science [20, 30, 19, 9, 35, 26], and it has been central to research in language [32, 25, 33, 36].”
A simple neural network module for relational reasoning (2017, background): “However, a number of these approaches, such as deep learning, often struggle in data-poor problems where the underlying structure is characterized by sparse but complex relations [3, 11].”
Robust Speech Recognition via Large-Scale Weak Supervision (2022, background): “…works have shown and continually reiterated how despite high performance on IID test sets, machine learning models can still make many mistakes when evaluated in even slightly different settings (Lake et al., 2017; Jia & Liang, 2017; Alcorn et al., 2019; Barbu et al., 2019; Recht et al., 2019).”
Inferring and Executing Programs for Visual Reasoning (2017, background): “Such compositional reasoning is a hallmark of human intelligence, and allows people to solve a plethora of problems using a limited set of basic skills [27].”
Interaction Networks for Learning about Objects, Relations and Physics (2016, background): “Representing and reasoning about objects, relations and physics is a “core” domain of human common 17 sense knowledge [25], and among the most basic and important aspects of intelligence [27, 15].”
Compositional Attention Networks for Machine Reasoning (2018, background): “However, deep learning approaches often struggle to perform well on tasks with a compositional and structured nature (Garnelo et al., 2016; Lake et al., 2017).”
Toward Causal Representation Learning (2021, background): “Such a representation allows children to quickly learn new tasks as their knowledge and intuitive understanding of physics can be reused [17], [53], [144], [250].”
What do we need to build explainable AI systems for the medical domain? (2017, background): “From the perspective of fundamental research, the gained insights might contribute towards building machines that learn and think like people (Lake et al., 2015, 2016).”
Towards Deep Symbolic Reinforcement Learning (2016, background): “But the everyday physical world is structured according to many other common sense priors [28, 27, 29, 25, 30].”
Towards Causal Representation Learning (2021, background): “their knowledge and intuitive understanding of physics can be re-used [15, 52, 144, 250].”
Learning First-Order Symbolic Representations for Planning from the Structure of the State Space (2019, background): “Learners, and in particular deep learners, have achieved considerable success but result in black boxes that do not have the flexibility, transparency, and generality of their model-based counterparts [26, 27, 32, 12, 17].”
On the Transfer of Inductive Bias from Simulation to the Real World: a New Disentanglement Dataset (2019, background): “To bridge the gap between simulation and the physical world, we built a recording platform which allows to investigate the following research questions: (i) How well do unsupervised state-of-the-art algorithms transfer from rendered images to physical recordings? (ii) How much does this transfer dep”
Are Disentangled Representations Helpful for Abstract Visual Reasoning? (2019, background): “Learning good representations of high-dimensional sensory data is of fundamental importance to Artificial Intelligence [4, 3, 6, 49, 7, 69, 67, 50, 59, 73].”
Multi-Object Representation Learning with Iterative Variational Inference (2019, background): “Good disentanglement is believed to lead to both better generalization and more interpretable features (Lake et al., 2016; Higgins et al., 2017b).”