Lifeboat Foundation

The large language models that have increasingly taken over the tech world are not “cheap” in many ways. The most prominent LLMs, such as GPT-4, took some $100 million to build in the form of legal costs of accessing training data, computational power costs for what could be billions or trillions of parameters, the energy and water needed to fuel computation, and the many coders developing the training algorithms that must run cycle after cycle so the machine will “learn.”

But, if a researcher needs to do a specialized task that a machine could do more efficiently and they don’t have access to a large institution that offers access to generative AI tools, what other options are available? Say, a parent wants to prep their child for a difficult test and needs to show many examples of how to solve complicated math problems.

Building their own LLM is an onerous prospect for costs mentioned above, and making direct use of the big models like GPT-4 and Llama 3.1 might not immediately be suited for the complex reasoning in logic and math their task requires.

Model grounded in biology reveals the tissue structures linked to the disorder. A researcher’s mathematical modeling approach for brain imaging analysis reveals links between genes, brain structure and autism.

A multi-university research team co-led by University of Virginia engineering professor Gustavo K. Rohde has developed a system that can spot genetic markers of autism in brain images with 89 to 95% accuracy.

Their findings suggest doctors may one day see, classify and treat autism and related neurological conditions with this method, without having to rely on, or wait for, behavioral cues. And that means this truly personalized medicine could result in earlier interventions.

To expand its GPT capabilities, OpenAI released its long-anticipated o1 model, in addition to a smaller, cheaper o1-mini version. Previously known as Strawberry, the company says these releases can “reason through complex tasks and solve harder problems than previous models in science, coding, and math.”

Although it’s still a preview, OpenAI states this is the first of this series in ChatGPT and on its API, with more to come.

The company says these models have been training to “spend more time thinking through problems before they respond, much like a person would. Through training, they learn to refine their thinking process, try different strategies, and recognize their mistakes.”

This conversation between Max Tegmark and Joel Hellermark was recorded in April 2024 at Max Tegmark’s MIT office. An edited version was premiered at Sana AI Summit on May 15 2024 in Stockholm, Sweden.

Max Tegmark is a professor doing AI and physics research at MIT as part of the Institute for Artificial Intelligence \& Fundamental Interactions and the Center for Brains, Minds, and Machines. He is also the president of the Future of Life Institute and the author of the New York Times bestselling books Life 3.0 and Our Mathematical Universe. Max’s unorthodox ideas have earned him the nickname “Mad Max.”

Continue reading “Creating superhuman AI” »

This i share with pleasure H/T Ulla Mattfolk.

Physical insights drawn from the real world are inexplicably useful for solving abstruse problems in mathematics.

Mathematicians have described a new class of shape that characterizes forms commonly found in nature — from the chambers in the iconic spiral shell of the nautilus to the way in which seeds pack into plants.

‘Soft cells’ — shapes with rounded corners and pointed tips that fit together on a plane — feature in onions, molluscs and more.

With only 6.6B activate parameters, GRIN MoE achieves exceptionally good performance across a diverse set of tasks, particularly in coding and mathematics tasks.

Microsoft releases GRIN😁 MoE

GRadient-INformed MoE

Continue reading “Microsoft/GRIN-MoE” »

Scientists from the U.S. Department of Energy’s (DOE) Brookhaven National Laboratory have shown that a type of qubit whose architecture is more amenable to mass production can perform comparably to qubits currently dominating the field. With a series of mathematical analyses, the scientists have provided a roadmap for simpler qubit fabrication that enables robust and reliable manufacturing of these quantum computer building blocks.

In 1971, English mathematical physicist and Nobel-prize winner Roger Penrose proposed how energy could be extracted from a rotating black hole. He argued that this could be done by building a harness around the black hole’s accretion disk, where infalling matter is accelerated to close to the speed of light, triggering the release of energy in multiple wavelengths.

Since then, multiple researchers have suggested that advanced civilizations could use this method (the Penrose Process) to power their civilization and that this represents a technosignature we should be on the lookout for.

Examples include John M. Smart’s Transcension Hypothesis, a proposed resolution to the Fermi Paradox where he suggested advanced intelligence may migrate to the region surrounding black holes to take advantage of the energy available.