Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog

Category: robotics/AI – Page 1,449

A hyperparameter optimization library for reproducible research

The table also shows the average normalized rank of transfer learning approaches. Hyperparameter transfer learning uses evaluation data from past HPO tasks in order to warmstart the current HPO task, which can result in significant speed-ups in practice.

Syne Tune supports transfer-learning-based HPO via an abstraction that maps a scheduler and transfer learning data to a warmstarted instance of the former. We consider the bounding-box and quantile-based ASHA, respectively referred to as ASHA-BB and ASHA-CTS. We also consider a zero-shot approach (ZS), which greedily selects hyperparameter configurations that complement previously considered ones, based on historical performances; and RUSH, which warmstarts ASHA with the best configurations found for previous tasks. As expected, we find that transfer learning approaches accelerate HPO.

Our experiments show that Syne Tune makes research on automated machine learning more efficient, reliable, and trustworthy. By making simulation on tabulated benchmarks a first-class citizen, it makes hyperparameter optimization accessible to researchers without massive computation budgets. By supporting advanced use cases, such as hyperparameter transfer learning, it allows better problem solving in practice.

Pinpointing Consciousness in Animal Brain Using Mouse ‘Brain Map’

Summary: Brain mapping study identifies important neural networks and their connections that appear to enhance the conscious experience.

Source: University of Tokyo

Science may be one step closer to understanding where consciousness resides in the brain. A new study shows the importance of certain types of neural connections in identifying consciousness.

The research, published in Cerebral Cortex, was led by Jun Kitazono, a corresponding author and a project researcher in the Department of General Systems Studies at the University of Tokyo.

NASA greenlights two new Mars helicopters and lengthens Perseverance’s resume

NASA and the European Space Agency (ESA) have agreed to “significant and advantageous changes” to a major part of the conceptual design for its Perseverance mission, NASA associate administrator Thomas Zurburchen states in the recent announcement.

This car-sized rover is the newest member of NASA’s robotic Mars fleet, and reached the Red Planet in February 2021 through an unprecedented landing. Arguably one of its most important responsibilities is the Mars Sample Return campaign. Perseverance’s six wheels leave grooves on the planet’s regolith as it works towards that goal, traversing Mars’ Jezero Crater to gather the telltale sedimentary proof that water — and possibly life — once existed there.

In October, the space agencies will dive into the details of their redesign: rather than having Perseverance leave caches of its pebble collection on Mars’ surface for another yet-to-be-built land-based spacecraft to pick up, the existing Mars rover will be the one to carry the precious parcels to their launch site. In addition, Perseverance’s high-flying robotic companion, the Ingenuity helicopter, has inspired the design of two future rotorcraft that would swerve over the Martian terrain to pick up other samples. This duo would be part of an existing concept, NASA’s Sample Retrieval Lander.

DeepMind’s AI has now catalogued every protein known to science

In late 2020, Alphabet’s DeepMind division unveiled its novel protein fold prediction algorithm, AlphaFold, and helped solve a scientific quandary that had stumped researchers for half a century. In the year since its beta release, half a million scientists from around the world have accessed the AI system’s results and cited them in their own studies more than 4,000 times. On Thursday, DeepMind announced that it is increasing that access even further by radically expanding its publicly-available AlphaFold Protein Structure Database (AlphaFoldDB) — from 1 million entries to 200 million entries.

Alphabet partnered with EMBL’s European Bioinformatics Institute (EMBL-EBI) for this undertaking, which covers proteins from across the kingdoms of life — animal, plant, fungi, bacteria and others. The results can be viewed on the UniProt, Ensembl, and OpenTargets websites or downloaded individually via GitHub, “for the human proteome and for the proteomes of 47 other key organisms important in research and global health,” per the AlphaFold website.

“AlphaFold is the singular and momentous advance in life science that demonstrates the power of AI,” Eric Topol, Founder and Director of the Scripps Research Translational Institute, siad in a press statement Thursday. “Determining the 3D structure of a protein used to take many months or years, it now takes seconds. AlphaFold has already accelerated and enabled massive discoveries, including cracking the structure of the nuclear pore complex. And with this new addition of structures illuminating nearly the entire protein universe, we can expect more biological mysteries to be solved each day.”

A “Nano-Robot” Built Entirely from DNA to Explore Cell Processes

Constructing a tiny robot from DNA and using it to study cell processes invisible to the naked eye… You would be forgiven for thinking it is science fiction, but it is in fact the subject of serious research by scientists from Inserm, CNRS and Université de Montpellier at the Structural Biology Center in Montpellier[1]. This highly innovative “nano-robot” should enable closer study of the mechanical forces applied at microscopic levels, which are crucial for many biological and pathological processes. It is described in a new study published in Nature Communications.

Our cells are subject to mechanical forces exerted on a microscopic scale, triggering biological signals essential to many cell processes involved in the normal functioning of our body or in the development of diseases.

For example, the feeling of touch is partly conditional on the application of mechanical forces on specific cell receptors (the discovery of which was this year rewarded by the Nobel Prize in Physiology or Medicine).

Combining Neuroscience, Psychology, and AI Yields a Foundational Model of Human Thought

From Human to Artificial General Intelligence

Humans have an almost unbounded set of skills and knowledge, and quickly learn new information without needing to be re-engineered to do so. It is conceivable that an AGI can be built using an approach that is fundamentally different from human intelligence. However, as three longtime researchers in AI and cognitive science, our approach is to draw inspiration and insights from the structure of the human mind. We are working toward AGI by trying to better understand the human mind, and better understand the human mind by working toward AGI.

From research in neuroscience, cognitive science, and psychology, we know that the human brain is neither a huge homogeneous set of neurons nor a massive set of task-specific programs that each solves a single problem. Instead, it is a set of regions with different properties that support the basic cognitive capabilities that together form the human mind.