Infants outperform artificial intelligence in detecting what motivates other peopleโs actions, according to a new study.
The results, which highlight fundamental differences between cognition and computation, point to shortcomings in todayโs technologies and where improvements are needed for AI to more fully replicate human behavior.
โAdults and even infants can easily make reliable inferences about what drives other peopleโs actions,โ explains Moira Dillon, an assistant professor in New York Universityโs psychology department and the senior author of the paper in the journal Cognition. โCurrent AI finds these inferences challenging to make.โ
โThe novel idea of putting infants and AI head-to-head on the same tasks is allowing researchers to better describe infantsโ intuitive knowledge about other people and suggest ways of integrating that knowledge into AI,โ she adds.
โIf AI aims to build flexible, commonsense thinkers like human adults become, then machines should draw upon the same core abilities infants possess in detecting goals and preferences,โ says Brenden Lake, an assistant professor in NYUโs Center for Data Science and psychology department and one of the paperโs authors.
Itโs been well-established that infants are fascinated by other peopleโas evidenced by how long they look at others to observe their actions and to engage with them socially. In addition, previous studies focused on infantsโ โcommonsense psychologyโโtheir understanding of the intentions, goals, preferences, and rationality underlying othersโ actionsโhave indicated that infants are able to attribute goals to others and expect others to pursue goals rationally and efficiently. The ability to make these predictions is foundational to human social intelligence.
Conversely, โcommonsense AIโโdriven by machine-learning algorithmsโpredicts actions directly. This is why, for example, an ad touting San Francisco as a travel destination pops up on your computer screen after you read a news story on a newly elected city official. However, what AI lacks is flexibility in recognizing different contexts and situations that guide human behavior.
To develop a foundational understanding of the differences between humansโ and AIโs abilities, the researchers conducted a series of experiments with 11-month-old infants and compared their responses to those yielded by state-of-the-art learning-driven neural-network models.
To do so, they deployed the previously established โBaby Intuitions Benchmarkโ (BIB)โsix tasks probing commonsense psychology. BIB was designed to allow for testing both infant and machine intelligence, allowing for a comparison of performance between infants and machines and, significantly, providing an empirical foundation for building human-like AI.
Specifically, infants on Zoom watched a series of videos of simple animated shapes moving around the screenโsimilar to a video game. The shapesโ actions simulated human behavior and decision-making through the retrieval of objects on the screen and other movements. Similarly, the researchers built and trained learning-driven neural-network modelsโAI tools that help computers recognize patterns and simulate human intelligenceโand tested the modelsโ responses to the exact same videos.
Their results showed that infants recognize human-like motivations even in the simplified actions of animated shapes. Infants predict that these actions are driven by hidden but consistent goalsโfor example, the on-screen retrieval of the same object no matter what location itโs in and the movement of that shape efficiently even when the surrounding environment changes. Infants demonstrate such predictions through their longer looking to such events that violate their predictionsโa common and decades-old measurement for gauging the nature of infantsโ knowledge.
Adopting this โsurprise paradigmโ to study machine intelligence allows for direct comparisons between an algorithmโs quantitative measure of surprise and a well-established human psychological measure of surpriseโinfantsโ looking time. The models showed no such evidence of understanding the motivations underlying such actions, revealing that they are missing key foundational principles of commonsense psychology that infants possess.
โA human infantโs foundational knowledge is limited, abstract, and reflects our evolutionary inheritance, yet it can accommodate any context or culture in which that infant might live and learn,โ observes Dillon.
Support for the research came from the National Science Foundation and the Defense Advanced Projects Research Agency.
Source: New York University
The post Infants beat AI at detecting human motivations appeared first on Futurity.
The Department of Justice's consumer-protection branch has opened a criminal investigation into the conduct of Abbott Laboratories, one of the country's largest formula makers at the center of a contamination scandal and ongoing nationwide shortage.
The existence of the investigation was first reported by The Wall Street Journal. Though the DOJ is not commenting on it, a spokesperson for Abbott said the department has informed them of the investigation and that the company is "cooperating fully."
Federal regulators last year found numerous violations and "egregiously unsanitary" conditions at Abbott's Sturgis, Michigan, plant, the largest formula factory in the country. The regulators previously received reports that at least four babies who drank formula made at that facility fell ill with dangerous infections of the bacterium Cronobacter sakazakii, which had also been detected in the plant. Two of the infants died.