Self-supervised learning through the eyes of a child
- URL: http://arxiv.org/abs/2007.16189v3
- Date: Tue, 15 Dec 2020 18:24:16 GMT
- Title: Self-supervised learning through the eyes of a child
- Authors: A. Emin Orhan, Vaibhav V. Gupta, Brenden M. Lake
- Abstract summary: We show the emergence of powerful, high-level visual representations from developmentally realistic natural videos using generic self-supervised learning objectives.
Our results demonstrate the emergence of powerful, high-level visual representations from developmentally realistic natural videos.
- Score: 35.235974685889396
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Within months of birth, children develop meaningful expectations about the
world around them. How much of this early knowledge can be explained through
generic learning mechanisms applied to sensory data, and how much of it
requires more substantive innate inductive biases? Addressing this fundamental
question in its full generality is currently infeasible, but we can hope to
make real progress in more narrowly defined domains, such as the development of
high-level visual categories, thanks to improvements in data collecting
technology and recent progress in deep learning. In this paper, our goal is
precisely to achieve such progress by utilizing modern self-supervised deep
learning methods and a recent longitudinal, egocentric video dataset recorded
from the perspective of three young children (Sullivan et al., 2020). Our
results demonstrate the emergence of powerful, high-level visual
representations from developmentally realistic natural videos using generic
self-supervised learning objectives.
Related papers
- Open-World Reinforcement Learning over Long Short-Term Imagination [91.28998327423295]
We present LS-Imagine, which extends the imagination horizon within a limited number of state transition steps.
Our method demonstrates significant improvements over state-of-the-art techniques in MineDojo.
arXiv Detail & Related papers (2024-10-04T17:17:30Z) - The BabyView dataset: High-resolution egocentric videos of infants' and young children's everyday experiences [8.952954042940368]
We release the largest developmental egocentric video dataset to date -- the BabyView dataset.
This 493 hour dataset includes egocentric videos from children spanning 6 months - 5 years of age in both longitudinal, at-home contexts and in a preschool environment.
We train self-supervised language and vision models and evaluate their transfer to out-of-distribution tasks including syntactic structure learning, object recognition, depth estimation, and image segmentation.
arXiv Detail & Related papers (2024-06-14T23:52:27Z) - Self-supervised learning of video representations from a child's perspective [27.439294457852423]
Children learn powerful internal models of the world around them from a few years of egocentric visual experience.
Can such internal models be learned from a child's visual experience with highly generic learning algorithms or do they require strong inductive biases?
arXiv Detail & Related papers (2024-02-01T03:27:26Z) - A Comprehensive Survey of Continual Learning: Theory, Method and
Application [64.23253420555989]
We present a comprehensive survey of continual learning, seeking to bridge the basic settings, theoretical foundations, representative methods, and practical applications.
We summarize the general objectives of continual learning as ensuring a proper stability-plasticity trade-off and an adequate intra/inter-task generalizability in the context of resource efficiency.
arXiv Detail & Related papers (2023-01-31T11:34:56Z) - Intrinsically Motivated Learning of Causal World Models [0.0]
A promising direction is to build world models capturing the true physical mechanisms hidden behind the sensorimotor interaction with the environment.
Inferring the causal structure of the environment could benefit from well-chosen actions as means to collect relevant interventional data.
arXiv Detail & Related papers (2022-08-09T16:48:28Z) - Deep Learning to See: Towards New Foundations of Computer Vision [88.69805848302266]
This book criticizes the supposed scientific progress in the field of computer vision.
It proposes the investigation of vision within the framework of information-based laws of nature.
arXiv Detail & Related papers (2022-06-30T15:20:36Z) - Deep Long-Tailed Learning: A Survey [163.16874896812885]
Deep long-tailed learning aims to train well-performing deep models from a large number of images that follow a long-tailed class distribution.
Long-tailed class imbalance is a common problem in practical visual recognition tasks.
This paper provides a comprehensive survey on recent advances in deep long-tailed learning.
arXiv Detail & Related papers (2021-10-09T15:25:22Z) - The Next Big Thing(s) in Unsupervised Machine Learning: Five Lessons
from Infant Learning [2.9005223064604078]
We argue that developmental science of infant cognition might hold the key to unlocking the next generation of unsupervised learning approaches.
Human infant learning is the closest biological parallel to artificial unsupervised learning, as infants too must learn useful representations from unlabelled data.
We identify five crucial factors enabling infants' quality and speed of learning, assess the extent to which these have already been exploited in machine learning, and propose how further adoption of these factors can give rise to previously unseen performance levels in unsupervised learning.
arXiv Detail & Related papers (2020-09-17T18:47:06Z) - Self-supervised Learning: Generative or Contrastive [16.326494162366973]
Self-supervised learning has soaring performance on representation learning in the last several years.
We take a look into new self-supervised learning methods for representation in computer vision, natural language processing, and graph learning.
arXiv Detail & Related papers (2020-06-15T08:40:03Z) - A Developmental Neuro-Robotics Approach for Boosting the Recognition of
Handwritten Digits [91.3755431537592]
Recent evidence shows that a simulation of the children's embodied strategies can improve the machine intelligence too.
This article explores the application of embodied strategies to convolutional neural network models in the context of developmental neuro-robotics.
arXiv Detail & Related papers (2020-03-23T14:55:00Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.