Goto

Collaborating Authors

The Next Big Thing(s) in Unsupervised Machine Learning: Five Lessons from Infant Learning

arXiv.org Artificial Intelligence

After a surge in popularity of supervised Deep Learning, the desire to reduce the dependence on curated, labelled data sets and to leverage the vast quantities of unlabelled data available recently triggered renewed interest in unsupervised learning algorithms. Despite a significantly improved performance due to approaches such as the identification of disentangled latent representations, contrastive learning, and clustering optimisations, the performance of unsupervised machine learning still falls short of its hypothesised potential. Machine learning has previously taken inspiration from neuroscience and cognitive science with great success. However, this has mostly been based on adult learners with access to labels and a vast amount of prior knowledge. In order to push unsupervised machine learning forward, we argue that developmental science of infant cognition might hold the key to unlocking the next generation of unsupervised learning approaches. Conceptually, human infant learning is the closest biological parallel to artificial unsupervised learning, as infants too must learn useful representations from unlabelled data. In contrast to machine learning, these new representations are learned rapidly and from relatively few examples. Moreover, infants learn robust representations that can be used flexibly and efficiently in a number of different tasks and contexts. 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.


SEDRo: A Simulated Environment for Developmental Robotics

arXiv.org Artificial Intelligence

Even with impressive advances in application-specific models, we still lack knowledge about how to build a model that can learn in a human-like way and do multiple tasks. To learn in a human-like way, we need to provide a diverse experience that is comparable to humans. In this paper, we introduce our ongoing effort to build a simulated environment for developmental robotics (SEDRo). SEDRo provides diverse human experiences ranging from those of a fetus to a 12th-month-old. A series of simulated tests based on developmental psychology will be used to evaluate the progress of a learning model. We anticipate SEDRo to lower the cost of entry and facilitate research in the developmental robotics community.


Babies learn faster if they are on the same wavelength as their mothers, brain scans reveal

Daily Mail - Science & tech

Good mothers teach their babies best when they think along the same wavelength - literally. Scans have revealed that infants whose patterns of brain activity most closely resemble their mum's take on new information more quickly. Experts say that parents can encourage this'syncing' - known as neural synchrony - through social cues including smiling and maintaining eye contact. This mirroring of brain wave patterns was a good predictor of how well babies' learn about their environment, researhers say. A mother smiling or frowning at an object to express like or dislike influences a baby's brain activity and how they respond to a new toy, researchers found.


Lessons from infant learning for unsupervised machine learning - Nature Machine Intelligence

#artificialintelligence

The desire to reduce the dependence on curated, labeled datasets and to leverage the vast quantities of unlabeled data has triggered renewed interest in unsupervised (or self-supervised) learning algorithms. Despite improved performance due to approaches such as the identification of disentangled latent representations, contrastive learning and clustering optimizations, unsupervised machine learning still falls short of its hypothesized potential as a breakthrough paradigm enabling generally intelligent systems. Inspiration from cognitive (neuro)science has been based mostly on adult learners with access to labels and a vast amount of prior knowledge. To push unsupervised machine learning forward, we argue that developmental science of infant cognition might hold the key to unlocking the next generation of unsupervised learning approaches. We identify three crucial factors enabling infants’ quality and speed of learning: (1) babies’ information processing is guided and constrained; (2) babies are learning from diverse, multimodal inputs; and (3) babies’ input is shaped by development and active learning. We assess the extent to which these insights from infant learning have already been exploited in machine learning, examine how closely these implementations resemble the core insights, and propose how further adoption of these factors can give rise to previously unseen performance levels in unsupervised learning. Unsupervised machine learning algorithms reduce the dependence on curated, labeled datasets that are characteristic of supervised machine learning. The authors argue that the developmental science of infant cognition could inform the design of unsupervised machine learning approaches.


AI Predicts Autism Based on Infant Brain Scans

#artificialintelligence

Brain scans, analyzed using a type of artificial intelligence, can reveal whether 6-month-old babies are likely to develop autism, a new study shows. The study examined 59 infants who were at high risk of developing autism; that is, each had an older sibling with autism. The artificial intelligence predicted with 100 percent accuracy that 48 infants would not develop autism. In addition, of the 11 infants who did develop the disorder by the time they were 2 years old, the system correctly predicted nine of the cases. "It was extremely accurate," Robert Emerson, the lead author on the study and a former cognitive neuroscience postdoctoral fellow at the University of North Carolina (UNC), told Live Science.