Humans learn differently than machines face_with_colon_three
This paper introduces ‘prospective configuration’, a new principle for learning in neural networks, which differs from backpropagation and is more efficient in learning and more consistent with data on neural activity and behavior.
But deep learning has a massive drawback: The algorithms can’t justify their answers. Often called the “black box” problem, this opacity stymies their use in high-risk situations, such as in medicine. Patients want an explanation when diagnosed with a life-changing disease. For now, deep learning-based algorithms—even if they have high diagnostic accuracy—can’t provide that information.
To open the black box, a team from the University of Texas Southwestern Medical Center tapped the human mind for inspiration. In a study in Nature Computational Science, they combined principles from the study of brain networks with a more traditional AI approach that relies on explainable building blocks.
The resulting AI acts a bit like a child. It condenses different types of information into “hubs.” Each hub is then transcribed into coding guidelines for humans to read—CliffsNotes for programmers that explain the algorithm’s conclusions about patterns it found in the data in plain English. It can also generate fully executable programming code to try out.
Automated algorithm discovery has been difficult for artificial intelligence given the immense search space of possible functions. Here explainable neural networks are used to discover algorithms that outperform those designed by humans.
Scientists are experimenting to figure out how best to attach ‘living skin’ to robotic faces, and make them smile. The results are, predictably, terrifying.
How human-like is artificial intelligence? In this blog post, we explore the consciousness, awareness, and intelligence of AI.
Designing machines to think like humans provides insight into intelligence itself.
The dream of artificial intelligence has never been just to make a grandmaster-beating chess engine or a chatbot that tries to break up a marriage. It has been to hold a mirror to our own intelligence, that we might understand ourselves better. Researchers seek not simply artificial intelligence but artificial general intelligence, or AGI—a system with humanlike adaptability and creativity.
Posted in genetics, robotics/AI
Building a conscious robot is a grand scientific and technological challenge. Debates about the possibility of conscious robots and the related positive outcomes and hazards for human beings are today no more confined to philosophical circles. Robot consciousness is a research field aimed to a unified view of approaches as cognitive robotics, epigenetic and affective robotics, situated and embodied robotics, developmental robotics, anticipatory systems, biomimetic robotics. Scholars agree that a conscious robot would completely change the current views on technology: it would not be an “intelligent companion” but a complete novel kind of artifact. Notably, many neuroscientists involved in the study of consciousness do not exclude this possibility. Moreover, facing the problem of consciousness in robots may be a major move on the study of consciousness in humans and animals.
Deducing the correct pattern that links pairs of coloured grids is relatively easy for most people, but relies on skills that artificial intelligence models lack. A new $1 million prize hopes to encourage the development of an AI that can solve such puzzles.
By Alex Wilkins
Scientists are using artificial intelligence (AI) to identify new animal species. But can we trust the results?
For now, scientists are using AI just to flag potentially new species; highly specialized biologists still need to formally describe those species and decide where they fit on the evolutionary tree. AI is also only as good as the data we train it on, and at the moment, there are massive gaps in our understanding of Earth’s wildlife.
Posted in robotics/AI
