Two Is Harder To Recognize Than Tom: the Challenge of Visual Numerosity for Deep Learning

02/09/2018
by   Xiaolin Wu, et al.
0

In the spirit of Turing test, we design and conduct a set of visual numerosity experiments with deep neural networks. We train DCNNs with a large number of sample images that are varied visual representations of small natural numbers, towards the objective of learning numerosity perception. Numerosity perception, or the number sense, is a cognitive construct so primary and so critical to the survival and well-being of our species that is considered and proven to be innate to human infants, and it responds to visual stimuli prior to the development of any symbolic skills, language or arithmetic. Somewhat surprisingly, in our experiments, even with strong supervision, DCNNs cannot see through superficial variations in visual representations and distill the abstract notion of natural number, a task that children perform with high accuracy and confidence. DCNNs are apparently easy to be confused by geometric variations and fail to grasp the topological essence in numerosity. The failures of DCNNs in the proposed cognition experiments also expose their overreliance on sample statistics at the expense of image semantics. Our findings are, we believe, significant and thought-provoking in the interests of AI research, because visual-based numerosity is a benchmark of minimum sort for human intelligence.

READ FULL TEXT

page 3

page 4

page 5

page 7

page 8

research
04/25/2020

Machine Number Sense: A Dataset of Visual Arithmetic Problems for Abstract and Relational Reasoning

As a comprehensive indicator of mathematical thinking and intelligence, ...
research
10/25/2021

IconQA: A New Benchmark for Abstract Diagram Understanding and Visual Language Reasoning

Current visual question answering (VQA) tasks mainly consider answering ...
research
10/25/2018

Decoding Brain Representations by Multimodal Learning of Neural Activity and Visual Features

This paper tackles the problem of learning brain-visual representations ...
research
10/12/2017

Can the early human visual system compete with Deep Neural Networks?

We study and compare the human visual system and state-of-the-art deep n...
research
09/13/2018

Sequential Coordination of Deep Models for Learning Visual Arithmetic

Achieving machine intelligence requires a smooth integration of percepti...
research
05/22/2022

Blackbird's language matrices (BLMs): a new benchmark to investigate disentangled generalisation in neural networks

Current successes of machine learning architectures are based on computa...
research
10/02/2018

Characterization of Visual Object Representations in Rat Primary Visual Cortex

For most animal species, quick and reliable identification of visual obj...

Please sign up or login with your details

Forgot password? Click here to reset