How modular should neural module networks be for systematic generalization?
V D'Amario, T Sasaki, X Boix - Advances in Neural …, 2021 - proceedings.neurips.cc
Abstract Neural Module Networks (NMNs) aim at Visual Question Answering (VQA) via
composition of modules that tackle a sub-task. NMNs are a promising strategy to achieve …
composition of modules that tackle a sub-task. NMNs are a promising strategy to achieve …
[HTML][HTML] Three approaches to facilitate invariant neurons and generalization to out-of-distribution orientations and illuminations
The training data distribution is often biased towards objects in certain orientations and
illumination conditions. While humans have a remarkable capability of recognizing objects …
illumination conditions. While humans have a remarkable capability of recognizing objects …
Deephys: Deep electrophysiology, debugging neural networks under distribution shifts
Deep Neural Networks (DNNs) often fail in out-of-distribution scenarios. In this paper, we
introduce a tool to visualize and understand such failures. We draw inspiration from …
introduce a tool to visualize and understand such failures. We draw inspiration from …
[PDF][PDF] To which out-of-distribution object orientations are dnns capable of generalizing
Abstract The capability of Deep Neural Networks (DNNs) to recognize objects in orientations
outside the distribution of the training data, ie., out-of-distribution (OoD) orientations, is not …
outside the distribution of the training data, ie., out-of-distribution (OoD) orientations, is not …
Three approaches to facilitate DNN generalization to objects in out-of-distribution orientations and illuminations
The training data distribution is often biased towards objects in certain orientations and
illumination conditions. While humans have a remarkable capability of recognizing objects …
illumination conditions. While humans have a remarkable capability of recognizing objects …
Emergent Neural Network Mechanisms for Generalization to Objects in Novel Orientations
The capability of Deep Neural Networks (DNNs) to recognize objects in orientations outside
the distribution of the training data is not well understood. We present evidence that DNNs …
the distribution of the training data is not well understood. We present evidence that DNNs …