A Perspective on Objects and Systematic Generalization in Model-Based RL

By Sjoerd van Steenkiste, Klaus Greff, J. Urgen Schmidhuber. 2019

31 citations recorded by the source corpus.

In order to meet the diverse challenges in solving many real-world problems, an intelligent agent has to be able to dynamically construct a model of its environment. Objects facilitate the modular reuse of prior knowledge and the combinatorial construction of such models. In this work, we argue that dynamically bound features (objects) do not simply emerge in connectionist models of the world. We identify several requirements that need to be fulfilled in overcoming this limitation and highlight corresponding inductive biases.

Subfields: Artificial Intelligence, Computer Networks and Communications, Reinforcement Learning

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 Computer Networks and Communications

In Reinforcement Learning

Citations in the corpus

Enhancing the Transformer with Explicit Relational Encoding for Math Problem Solving (2019, background): “…to enable the fully distributed representation of symbolic structure (like a hierarchy of features) which has recently resurfaced as an important direction for neural network research (Lake & Baroni, 2017; Bahdanau et al., 2018; van Steenkiste et al., 2019; Palangi et al., 2017; Tang et al., 2018).”
On the Binding Problem in Artificial Neural Networks (2020, background): “** A preliminary version of this work was presented at an ICML Workshop (van Steenkiste et al., 2019a).”
Hierarchical Relational Inference (2020, methodology): “In our case, we adopt an approach based on spatial slots (van Steenkiste, Greff, and Schmidhuber 2019; Greff, van Steenkiste, and Schmidhuber 2020) that puts less emphasis on the exact correspondence to objects.”
Remix: Rebalanced Mixup (2020, background): “age classication 1 Introduction Deep neural networks have made notable breakthroughs in many elds such as computer vision [35,27,20], natural language processing [18,19,9] and reinforcement learning [28]. Aside from delicately-designed algorithms and architectures, training data is one of the critic”
R-SQAIR: Relational Sequential Attend, Infer, Repeat (2019, background): “To facilitate the solution of real-world problems, intelligent agents should be able to acquire such knowledge [31].”
Unsupervised Object Keypoint Learning using Local Spatial Predictability (2020, background): “Similarly, when explicitly distinguishing abstract concepts, such as objects, at a representational level, it is easier to generalize (systematically) to novel scenes that are composed of these same abstract building blocks (Lake et al., 2017; van Steenkiste et al., 2019).”
Generative Neurosymbolic Machines (2020, background): “These properties, however, are believed to be crucial in resolving limitations of current deep learning in various System 2 [26] related abilities such as reasoning [4], causal learning [37, 34], accountability [11], and systematic out-of-distribution generalization [2, 42].”