AI Learns Language Like a Child

Summary: Researchers made a significant breakthrough by training a multimodal AI system using only the input one child received from birth through their second birthday, challenging the notion that AI requires vast data to learn language.

Their study demonstrates that the AI model was able to learn words and concepts from a fraction of a child’s experiences, captured through headcam recordings. This experiment highlights the potential of AI to mimic human language learning processes and reshapes our understanding of early language and concept acquisition.

By aligning AI learning with a child’s naturalistic experience, the researchers offer new insights into the debate on how children learn language, suggesting that associative learning may play a more substantial role than previously thought.

Key Facts:

  1. The AI system trained on headcam footage from a single child managed to learn a significant number of words and concepts, despite the video capturing only about 1% of the child’s waking hours.
  2. The study utilized a multimodal neural network, combining visual and linguistic data through contrastive learning, to mimic the way children link words with visual contexts.
  3. This research challenges traditional beliefs about language learning, indicating that associative learning with minimal input can lead to substantial language acquisition, much like in human children.

Source: NYU

AI systems, such as GPT-4, can now learn and use human language, but they learn from astronomical amounts of language input—much more than children receive when learning how to understand and speak a language. The best AI systems train on text with a word count in the trillions, whereas children receive just millions per year.

Due to this enormous data gap, researchers have been skeptical that recent AI advances can tell us much about human learning and development. An ideal test for demonstrating a connection would involve training an AI model, not on massive data from the web, but on only the input that a single child receives. What would the model be able to learn then?  

Credit: NYU

A team of New York University researchers ran this exact experiment. They trained a multimodal AI system through the eyes and ears of a single child, using headcam video recordings from when the child was six months and through their second birthday. They examined if the AI model could learn words and concepts present in a child’s everyday experience.

Their findings, reported in the latest issue of the journal Science, showed that the model, or neural network, could, in fact, learn a substantial number of words and concepts using limited slices of what the child experienced. That is, the video only captured about 1% of the child’s waking hours, but that was sufficient for genuine language learning. 

“We show, for the first time, that a neural network trained on this developmentally realistic input from a single child can learn to link words to their visual counterparts,” says Wai Keen Vong, a research scientist at NYU’s Center for Data Science and the paper’s first author.

“Our results demonstrate how recent algorithmic advances paired with one child’s naturalistic experience has the potential to reshape our understanding of early language and concept acquisition.”

“By using AI models to study the real language-learning problem faced by children, we can address classic debates about what ingredients children need to learn words—whether they need language-specific biases, innate knowledge, or just associative learning to get going,” adds Brenden Lake, an assistant professor in NYU’s Center for Data Science and Department of Psychology and the paper’s senior author.

“It seems we can get more with just learning than commonly thought.”

This shows a child and a robot.
For instance, when a parent says something in view of the child, it is likely that some of the words used are likely referring to something that the child can see, meaning comprehension is instilled by linking visual and linguistic cues. Credit: Neuroscience News

Vong, Lake, and their NYU colleagues, Wentao Wang and Emin Orhan, analyzed a child’s learning process captured on first-person video—via a light, head-mounted camera—on a weekly basis beginning at six months and through 25 months, using more than 60 hours of footage.

The footage contained approximately a quarter of a million word instances (i.e., the number of words communicated, many of them repeatedly) that are linked with video frames of what the child saw when those words were spoken and included a wide range of different activities across development, including mealtimes, reading books, and the child playing.

The NYU researchers then trained a multimodal neural network with two separate modules: one that takes in single video frames (the vision encoder) and another that takes in the transcribed child-directed speech (the language encoder).

These two encoders were combined and trained using an algorithm called contrastive learning, which aims to learn useful input features and their cross-modal associations. For instance, when a parent says something in view of the child, it is likely that some of the words used are likely referring to something that the child can see, meaning comprehension is instilled by linking visual and linguistic cues.

“This provides the model a clue as to which words should be associated with which objects,” explains Vong.

“Combining these cues is what enables contrastive learning to gradually determine which words belong with which visuals and to capture the learning of a child’s first words.”

After training the model, the researchers tested it using the same kinds of evaluations used to measure word learning in infants—presenting the model with the target word and an array of four different image options and asking it to select the image that matches the target word.

Their results showed that the model was able to learn a substantial number of the words and concepts present in the child’s everyday experience. Furthermore, for some of the words the model learned, it could generalize them to very different visual instances than those seen at training, reflecting an aspect of generalization also seen in children when they are tested in the lab.

“These findings suggest that this aspect of word learning is feasible from the kind of naturalistic data that children receive while using relatively generic learning mechanisms such as those found in neural networks,” observes Lake. 

Funding: The work was supported by the U.S. Department of Defense’s Defense Advanced Research Projects Agency (N6600119C4030) and the National Science Foundation (1922658). Participation of the child was approved by the parents and the methodology was approved by NYU’s Institutional Review Board.

About this artificial intelligence research news

Author: James Devitt
Source: NYU
Contact: James Devitt – NYU
Image: The image is credited to Neuroscience News

Original Research: Closed access.
Grounded language acquisition through the eyes and ears of a single child” by Wai Keen Vong et al. Science


Grounded language acquisition through the eyes and ears of a single child

Starting around 6 to 9 months of age, children begin acquiring their first words, linking spoken words to their visual counterparts. How much of this knowledge is learnable from sensory input with relatively generic learning mechanisms, and how much requires stronger inductive biases?

Using longitudinal head-mounted camera recordings from one child aged 6 to 25 months, we trained a relatively generic neural network on 61 hours of correlated visual-linguistic data streams, learning feature-based representations and cross-modal associations.

Our model acquires many word-referent mappings present in the child’s everyday experience, enables zero-shot generalization to new visual referents, and aligns its visual and linguistic conceptual systems.

These results show how critical aspects of grounded word meaning are learnable through joint representation and associative learning from one child’s input.

Join our Newsletter
I agree to have my personal information transferred to AWeber for Neuroscience Newsletter ( more information )
Sign up to receive our recent neuroscience headlines and summaries sent to your email once a day, totally free.
We hate spam and only use your email to contact you about newsletters. You can cancel your subscription any time.