Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog

Category: mathematics – Page 207

Neural networks explained

In the past 10 years, the best-performing artificial-intelligence systems—such as the speech recognizers on smartphones or Google’s latest automatic translator—have resulted from a technique called “deep learning.”

Deep learning is in fact a new name for an approach to artificial intelligence called neural networks, which have been going in and out of fashion for more than 70 years. Neural networks were first proposed in 1944 by Warren McCullough and Walter Pitts, two University of Chicago researchers who moved to MIT in 1952 as founding members of what’s sometimes called the first cognitive science department.

Neural nets were a major area of research in both neuroscience and computer science until 1969, when, according to computer science lore, they were killed off by the MIT mathematicians Marvin Minsky and Seymour Papert, who a year later would become co-directors of the new MIT Artificial Intelligence Laboratory.

Affordable deep space missions using asteroids

Phase 1 work demonstrated Optical Mining in the laboratory and performed mission and systems analysis of the application of Optical Mining to human exploration missions. Their mission analysis showed that the most accessible Near Earth Objects (NEOs) can be used to provide NASA with mission consumables for human exploration in deep space with the potential of saving up to $10 billion per year or $150 billion over the 15 year operational life cycle of a human exploration program. This savings alone would be enough to transform NASA’s vision of human exploration from being unaffordable to being affordable within budgets that Congress can approve. Phase 1 technical work included a full scale (8 kW) Optical Mining demonstration using a high fidelity CI-type asteroid simulant in vacuum using sunlight from a 10 meter diameter solar concentrator without mechanical contact or downforce. This work confirmed our physics based mathematical model of the excavation and volatile extraction process and scalability of results from 36 prior, small scale (≈ 1 cm diameter) demonstrations and tests.

Phase 2 work will complete mission and system analysis of the application of Optical Mining to an exciting program of human exploration and we will mature the technology of Optical Mining to the point at which NASA can baseline this approach for an affordable program of human exploration. Our mission studies will address the production via Optical Mining missions to extract and retrieve resources, consumable processing, storage, and application of consumables to human exploration mission in cislunar, NEO and Martian space. The mission studies will be tightly coupled with our laboratory work. Laboratory work will include the development and integration of a 30 kW Optical Mining test apparatus in our laboratory and integration with our high quality vacuum chamber for a test program involving Optical Mining.

We May Be Able to Build a Rocket That Can Go 99.999% the Speed of Light

Theoretical physics often lifts the sanctions we set on our own imaginations. Whether it’s exploring the possibility of warp drives or understanding the rate of the universe’s expansion, we are quick to explore the unknown on our chalkboards until our tech is ready for our ideas.

In a similar deep-dive into the theoretical, a Norwegian professor argues in the journal Acta Astronautica for the of possibility of photon rockets that can reach 99.999 percent of the speed of light (300,000 km/s [186,000 mph]); asserting that, while humanity can’t do it anytime soon, we could potentially build a spacecraft that falls just short of the ultimate speed limit sometime in the future when the necessary technology is feasible.

*2* A Finance Professor Predicts the Absolute Speed Limit for all Human Spacecrafts

Why Mary Lou Jepsen Left Facebook: To Transform Health Care and Invent Consumer Telepathy

“Yep, transforming health care and telepathy, those are the items on her to-do list. Jepsen plans to achieve both goals with a cheap wearable device that her engineers are now tinkering with in the lab. And then there’s the side benefit of reinvigorating the tired consumer electronics industry, which Jepsen thinks is due for the next big thing.

Jepsen was at SXSW to give a talk about Openwater, her new startup. While the company is still conducting R&D to decide on its first products, Jepsen feels the need to speak out now about what she’s building and how she thinks her technology could radically change society. She wants to give people fair warning and time to think about what’s coming. “I know it seems outlandish to be talking about telepathy, but it’s completely solid physics and mathematical principles—it’s in reach in the next three years,” she says.

Plus, she’s sick of stealth mode. “I haven’t been able to to talk about what I’ve been doing for five and half years while I was at Google and Facebook, and I don’t think secrecy is useful,” she says. She left Facebook in August, and in September she filed patents for her Openwater technology, which she expects to be issued any day now.

Can math help explain our bodies—and our diseases?

What makes a cluster of cells become a liver, or a muscle? How do our genes give rise to proteins, proteins to cells, and cells to tissues and organs?

The incredible complexity of how these biological systems interact boggles the mind—and drives the work of biomedical scientists around the world.

But a pair of mathematicians has introduced a new way of thinking about these concepts that may help set the stage for better understanding of our bodies and other living things.

Hard drives of the future could be made of DNA

Our data-driven society is churning out more information than traditional storage technology can handle, so scientists are looking for a solution in Nature’s hard drive: DNA. A pair of researchers at Columbia University and the New York Genome Center recently wrote a full computer operating system, an 1895 French film, an Amazon gift card and other files into DNA strands and retrieved them without errors, according to a study published in the latest edition of Science.

There are several advantages to using DNA. It’s a lot smaller than traditional media; a single gram can fit 215,000 times more data than a one terabyte hard drive, The Atlantic notes. It’s also incredibly durable. Scientists are using DNA thousands of years old to de-extinct wooly mammoths, for example. But, until now, they’ve only unlocked a fraction of its storage capacity. Study coauthors Yaniv Erlich and Dina Zielinski were able to fit the theoretical maximum amount of information per nucleotide using a new method inspired by how movies stream across the internet.

“We mapped the bits of the files to DNA nucleotides. Then, we synthesized these nucleotides and stored the molecules in a test-tube,” Erlich explained in an interview with ResearchGate. “To retrieve the information, we sequenced the molecules. This is the basic process. To pack the information, we devised a strategy—called DNA Fountain—that uses mathematical concepts from coding theory. It was this strategy that allowed us to achieve optimal packing, which was the most challenging aspect of the study.”

Playing favorites: Brain cells prefer one parent’s gene over the other’s

Well, in my immediate family; we get science, math, and futurists talents from my dad. And, there does seem to be a pattern in my immediate family with this; not sure about others. Would love to know though.

SALT LAKE CITY — Most kids say they love their mom and dad equally, but there are times when even the best prefers one parent over the other. The same can be said for how the body’s cells treat our DNA instructions. It has long been thought that each copy — one inherited from mom and one from dad — is treated the same. A new study from scientists at the University of Utah School of Medicine shows that it is not uncommon for cells in the brain to preferentially activate one copy over the other. The finding breaks basic tenants of classic genetics and suggests new ways in which genetic mutations might cause brain disorders.

In at least one region of the newborn mouse brain, the new research shows, inequality seems to be the norm. About 85 percent of genes in the dorsal raphe nucleus, known for secreting the mood-controlling chemical serotonin, differentially activate their maternal and paternal gene copies. Ten days later in the juvenile brain, the landscape shifts, with both copies being activated equally for all but 10 percent of genes.

More than an oddity of the brain, the disparity also takes place at other sites in the body, including liver and muscle. It also occurs in humans.

/* */