Re: AI & Robotics News and Discussions
Posted: Tue May 02, 2023 9:24 am
A community of futurology enthusiasts
https://www.futuretimeline.net/forum/
Researchers have created atomically thin artificial neurons capable of processing both light and electric signals for computing. The material enables the simultaneous existence of separate feedforward and feedback paths within a neural network, boosting the ability to solve complex problems.
For decades, scientists have been investigating how to recreate the versatile computational capabilities of biological neurons to develop faster and more energy-efficient machine learning systems. One promising approach involves the use of memristors: electronic components capable of storing a value by modifying their conductance and then utilizing that value for in-memory processing.
However, a key challenge to replicating the complex processes of biological neurons and brains using memristors has been the difficulty in integrating both feedforward and feedback neuronal signals. These mechanisms underpin our cognitive ability to learn complex tasks, using rewards and errors.
A team of researchers at the University of Oxford, IBM Research Europe, and the University of Texas, have announced an important feat: the development of atomically thin artificial neurons created by stacking two-dimensional (2D) materials. The results have been published in Nature Nanotechnology.
In the study, the researchers expanded the functionality of the electronic memristors by making them responsive to optical as well as electrical signals. This enabled the simultaneous existence of separate feedforward and feedback paths within the network. The advancement allowed the team to create winner-take-all neural networks: computational learning programs with the potential for solving complex problems in machine learning, such as unsupervised learning in clustering and combinatorial optimization problems.
AMD MI300
A bombshell new report from Bloomberg states Microsoft is teaming up with Nvidia's nemesis—AMD—to ensure the company doesn't continue to fortify its position as the vanguard of AI technology. The two companies are reportedly working together on unspecified AI products that seek to rival Nvidia in the AI hardware and software market. Recently it was revealed that Microsoft has been working on its own AI chip it calls Athena for some time now. However, a Microsoft spokesperson has denied AMD's involvement in that project. That does not mean the two companies aren't working on a different AI chip that has never been revealed.
The (paywalled) Bloomberg article states Microsoft is providing financial support for AMD for its own expansion into AI products and that the two companies are also working on Athena together. Bloomberg says its sources include several people who are close to the project. If the sources are correct, it makes for a powerful pairing given Microsoft's existing $10 billion investment into OpenAI's ChatGPT, which it has integrated into Bing and its Edge browser. AMD has tremendous chip design assets but is much smaller than Nvidia and has always been the underdog in the industry.
https://markets.businessinsider.com/new ... gn=tech-sfMay 4, 2023, 4:31 PM BST
A theoretical stock portfolio of 38 companies picked by ChatGPT has outperformed the 10 most popular funds in the UK over an eight week stretch, according to finance website Finder.com.
The AI-powered fund has climbed 4.93% in the eight weeks since March 6, 2023, while the group of competing funds together averaged a loss of 0.78% in the same period. The bot-picked fund has outperformed the real funds in 34 of the 39 market days the experiment has been running, the report said.
The top three performing stocks in the ChatGPT portfolio were Meta, Microsoft, and Intel Corporation, which saw increases of roughly 30%, 20%, and 18%, respectively.
Other names that rounded out the portfolio include Visa, Home Depot, Johnson & Johnson, Nvidia, and Netflix.
Finder asked the AI large language model to pick stocks using a range of investing principles taken from the leading funds, the website said, and the study bypassed ChatGPT's warning that it "cannot provide specific investment advice" by adding that it was a theoretical exercise.
Thomas Edison once described genius as "one percent inspiration and 99 percent perspiration" – but AI systems like MetaGPT can already reduce that sweat to nearly nothing when it comes to coding and deploying simple apps and websites.
Large Language Models (LLMs) can "understand" natural language prompts with incredible insight and subtlety, and they can write highly effective code in several programming languages, too. But in their raw question-and-answer form – such as ChatGPT – they don't exactly just go away and get the job done for you.
That's why people are building a million apps around these AIs, constraining their vast abilities into single-function machines. An app built around a single type of task can do a lot of background work preparing and assisting the AI in its job, and trying to ensure a high quality output.