Lifeboat Foundation

In the field of artificial intelligence, reinforcement learning is a type of machine-learning strategy that rewards desirable behaviors while penalizing those which aren’t. An agent can perceive its surroundings and act accordingly through trial and error in general with this form or presence – it’s kind of like getting feedback on what works for you. However, learning rules from scratch in contexts with complex exploration problems is a big challenge in RL. Because the agent does not receive any intermediate incentives, it cannot determine how close it is to complete the goal. As a result, exploring the space at random becomes necessary until the door opens. Given the length of the task and the level of precision required, this is highly unlikely.

Exploring the state space randomly with preliminary information should be avoided while performing this activity. This prior knowledge aids the agent in determining which states of the environment are desirable and should be investigated further. Offline data collected by human demonstrations, programmed policies, or other RL agents could be used to train a policy and then initiate a new RL policy. This would include copying the pre-trained policy’s neural network to the new RL policy in the scenario where we utilize neural networks to describe the procedures. This process transforms the new RL policy into a pre-trained one. However, as seen below, naively initializing a new RL policy like this frequently fails, especially for value-based RL approaches.

Google AI researchers have developed a meta-algorithm to leverage pre-existing policy to initialize any RL algorithm. The researchers utilize two procedures to learn tasks in Jump-Start Reinforcement Learning (JSRL): a guide policy and an exploration policy. The exploration policy is an RL policy trained online using the agent’s new experiences in the environment. In contrast, the guide policy is any pre-existing policy that is not modified during online training. JSRL produces a learning curriculum by incorporating the guide policy, followed by the self-improving exploration policy, yielding results comparable to or better than competitive IL+RL approaches.

Controllable protein attachment onto solid interfaces is essential for the functionality of proteins with broad applications. Silica-binding peptides (SBPs) have emerged as an important tool enabling convenient binding of proteins onto a silica surface. Surprisingly, we found that removal of polyhistidines, a common tag for protein purification, dramatically decrease the binding affinity of a SBP-tagged nanobody onto a silica surface. We hypothesized that polyhistidines and SBPs can be combined to enhance affinity. Through a series of purposely designed SBPs, we identified that the relative orientation of amino acids is a key factor affecting the surface binding strength. One re-engineered SBP, SBP4, exhibits a 4000-fold improvement compared to the original sequence.

Pulling up to a house in the suburbs of Boulder, Colorado, the white van seems unremarkable — almost nondescript, save for a few institutional decals. But when the door slides open, it reveals an unusual scene: the standard benches have been tossed out, traded for retractable tables, an examination chair, and a psychedelic wall rug (to mellow the vibe).

CSIS will host a public event on responsible AI in a global context, featuring a moderated discussion with Julie Sweet, Chair and CEO of Accenture, and Brad Smith, President and Vice Chair of the Microsoft Corporation, on the business perspective, followed by a conversation among a panel of experts on the best way forward for AI regulation. Dr. John J. Hamre, President and CEO of CSIS, will provide welcoming remarks.

Keynote Speakers:
Brad Smith, President and Vice Chair, Microsoft Corporation.
Julie Sweet, Chair and Chief Executive Officer, Accenture.

Continue reading “Responsible AI in a Global Context” »

NASA ’s citizen science projects are collaborations between scientists and interested members of the public. Through these collaborations, volunteers known as citizen scientists have helped make thousands of important scientific discoveries. Aurorasaurus is one such project that tracks auroras around the world in real time via reports on its website and on Twitter.

Aurorasaurus often partners with other organizations to complement science with citizen science and recently Aurorasauraus partnered with NASA’s Loss through Auroral Microburst Pulsations (LAMP) mission. Early on the morning of Saturday, March 5, 2022, the LAMP mission successfully took flight, flying straight into a pulsating aurora.

The ongoing success of the multi-instrument optics alignment for NASA ’s Webb telescope’s near-infrared instruments has moved the attention of the commissioning team to chill as we carefully monitor the cooling of the Mid-InfraRed Instrument (MIRI) down to its final operating temperature of less than 7 kelvins (−447 degrees Fahrenheit 0, or-266 degrees Celsius). We are continuing other activities during this slow cooldown which include monitoring the near-infrared instruments. As MIRI cools, other major components of the observatory, such as the backplane and mirrors, also continue to cool and are approaching their operational temperatures.

Last week, the Webb team did a station-keeping thruster burn to maintain Webb’s position in orbit around the second Lagrange point. This was the second burn since Webb’s arrival at its final orbit in January; these burns will continue periodically throughout the lifetime of the mission.

In the last few weeks, we have been sharing some of Webb’s anticipated science, beginning with the study of the first stars and galaxies in the early universe. Today, we will see how Webb will peer within our own Milky Way galaxy at places where stars and planets form. Klaus Pontoppidan, the Space Telescope Science Institute project scientist for Webbthe cool science planned for star and planet formation with Webb:

After a wild display of lights, music and futuristic technology, Tesla (TSLA) CEO Elon Musk kicked off the grand opening of the company’s new Texas gigafactory on Thursday.

The Austin plant — Tesla’s fourth globally — will manufacture the Model Y SUV and, next year, the highly-anticipated Cybertruck.

Continue reading “Tesla bull predicts between 5 to 10 new gigafactories in the next two years” »

🚘😏Roads?

☝️😉Where we’re going we don’t NEED roads!

ALI technologies’ president said the XTURISMO, which can reach a top speed of 80km/h, would prove popular in the Middle East.

But these particles would interact only weakly with ordinary matter, and only via the force of gravity. This description is eerily similar to what we know about dark matter, which does not interact with light yet has a gravitational influence felt everywhere in the universe. This gravitational influence, for instance, is what prevents galaxies from flying apart.

“The main advantage of massive gravitons as dark matter particles is that they only interact gravitationally, hence they can escape attempts to detect their presence,” Cacciapaglia said.

In contrast, other proposed dark matter candidates — such as weakly interacting massive particles, axions and neutrinos — might also be felt by their very subtle interactions with other forces and fields.