Most linkage systems today are designed manually because of the high level of precision needed. This makes the problem exceedingly challenging."
"But if you're designing a mechanical system, a small change may lead the whole thing to fail. "When you create an image using AI, you can get two pixels wrong, and it doesn't matter," said Faez Ahmed, a mechanical engineering professor at MIT.
Think of the wipers that clear rain and snow from your car windshield: A motor rotates an arm connected to links that move each wiper. Most moving machines require a linkage mechanism that transfers motion or force from one part to another. But try to design a windshield wiper that way, and it’s almost certain to fail. Show an AI model enough Impressionist art and it can learn to paint in that style. “Teaching the model an emergent language first can make it easier to learn non-Indo-European languages, while avoiding some of the cultural biases that come with pretraining on a Western language.” A new way of designing machines that move “It’s easier to learn German if you know English, but that’s not the case with non-Indo-European languages like Niger-Congo or Trans-New Guinea," said Chuang Gan, an IBM researcher with the MIT-IBM Watson AI Lab. They hypothesize that no matter what language we speak, our visual world is largely the same, creating a common foundation for natural language. In the end, the model performed nearly as well on a fill-in-the-blank fluency test as a model pretrained on Spanish, the researchers found. Trained on this prototype language, the model was fine-tuned on labeled text in Urdu, Basque, Persian and seven other languages. Eventually, an emergent language arises from these image-grounded symbols. A second algorithm compares the numbers to a set of images and picks the image that seems like the best match. One algorithm receives a visual prompt - like an image of a bedroom - and outputs a sequence of numbers. Researchers used a generative AI model to create some 2 million symbolic “tokens” in a game pairing symbols with natural images. “When humans learn to talk, they associate words with visual concepts,” said Yang Zhang, an IBM researcher with the MIT-IBM Watson AI Lab. In a paper spotlighted at ICLR this year, IBM researchers showed that pretraining a language model on a made-up language grounded in images could make it easier to master a low-resource language like Urdu. Thousands of spoken languages have relatively few texts in machine-readable form, stalling the development of AI applications for those languages. But it’s how IBM researchers are tackling the moonshot of developing AI applications for less dominant languages. Studying nonsense before trying to learn Urdu, Pakistan’s official language, might sound ridiculous. Here are five inventive ways that IBM is using synthetic data to improve AI models. “They can help to make AI models more fair, accurate, and trustworthy.” “Synthetic data can be an indispensable testing tool for AI,” said Inkit Padhi, an expert on synthetic data at IBM Research. Deployed as adversarial examples, fake data can show where an AI model is likely to make mistakes or unfair decisions. Synthetic data are also invaluable for probing a trained model for security flaws and biases. They also provide a low-cost stand-in for health care records, financial data, copyrighted material, and other data that are either highly regulated or come with privacy and ethical concerns. They’re commonly used to inject more variety into datasets to make machine-learning models more accurate and reliable. Today, synthetic data are everywhere, driving some of AI’s most innovative applications. Peele’s PSA on ‘deepfakes,’ audio and video manipulated to deceive, was the first time many people heard of synthetic data. “We're entering an era in which our enemies can make anyone say anything at any point in time.” In this viral video from 2018, actor-writer Jordan Peele projected his voice into former President Obama’s moving lips.
0 kommentar(er)
