PRESS-NEWS.org - Press Release Distribution
PRESS RELEASES DISTRIBUTION

Better paths yield better AI

Like climbing a mountain via the shortest possible path, improving classification tasks can be achieved by choosing the most influential path to the output, and not just by learning with deeper networks

Better paths yield better AI
2023-08-31
(Press-News.org)

Deep Learning (DL) performs classification tasks using a series of layers. To effectively execute these tasks, local decisions are performed progressively along the layers. But can we perform an all-encompassing decision by choosing the most influential path to the output rather than performing these decisions locally?

In an article published today in Scientific Reports, researchers from Bar-Ilan University in Israel answer this question with a resounding "yes". Pre-existing deep architectures have been improved by updating the most influential paths to the output.

"One can think of it as two children who wish to climb a mountain with many twists and turns. One of them chooses the fastest local route at every intersection while the other uses binoculars to see the entire path ahead and picks the shortest and most significant route, just like Google Maps or Waze. The first child might get a head start, but the second will end up winning," said Prof. Ido Kanter, of Bar-Ilan's Department of Physics and Gonda (Goldschmied) Multidisciplinary Brain Research Center, who led the research.

"This discovery can pave the way for better enhanced AI learning, by choosing the most significant route to the top,” added Yarden Tzach, a PhD student and one of the key contributors to this work.

This exploration of a deeper comprehension of AI systems by Prof. Kanter and his experimental research team, led by Dr. Roni Vardi, aims to bridge between the biological world and machine learning, thereby creating an improved, advanced AI system. To date they have discovered evidence for efficient dendritic adaptation using neuronal cultures, as well as how to implement those findings in machine learning, showing how shallow networks can compete with deep ones, and finding the mechanism underlying successful deep learning.

Enhancing existing architectures using global decisions can pave the way for improved AI, which can improve its classification tasks without the need for additional layers.

END


[Attachments] See images for this press release:
Better paths yield better AI Better paths yield better AI 2 Better paths yield better AI 3

ELSE PRESS RELEASES FROM THIS DATE:

Children’s books are still Whiter, and more male, than US society

2023-08-31
A new paper in the Quarterly Journal of Economics, published by Oxford University Press, finds that children’s books in the United States continue to underrepresent ethnic minorities. In addition, it finds that male characters are overrepresented in such stories and children are often presented with lighter skin tones for no apparent editorial reason. Education teaches children about the world, its people, and their place in it. Much of this happens through the books society presents to children ...

New insight for stabilizing halide perovskite via thiocyanate substitution

New insight for stabilizing halide perovskite via thiocyanate substitution
2023-08-31
α-FAPbI3, a promising solar cell material with a cubic perovskite structure that is metastable at room temperature, can be stabilized by introducing a pseudo-halide ion like thiocyanate (SCN–) into its structure, demonstrated by Tokyo Tech researchers in a new study. Their finding provides new insights into the stabilization of the α-phase via grain boundary and pseudo-halide engineering. The light we receive every day from the Sun, if harnessed efficiently, can help us tackle the ongoing global energy crisis as well as our concern with climate change. Materials with good photophysical properties, i.e., light absorption, ...

Scientists develop finger sweat test to detect antipsychotic drugs in patients

2023-08-31
Antipsychotic drugs treat incredibly vulnerable patients. Maintaining a treatment regimen is difficult for many patients, but not taking the medication is associated with a higher risk of poor health outcomes. These drugs are also very powerful with strong side-effects, and blood tests are often used to calibrate a patient’s dosage and confirm that they are taking the recommended dose.  However, blood tests are invasive and potentially uncomfortable. Scientists have now discovered a way to test the levels of common antipsychotic drugs in the sweat ...

Acting fast when an epidemic hits

2023-08-31
A team of researchers at the University of Waterloo and Dalhousie University have developed a method for forecasting the short-term progression of an epidemic using extremely limited amounts of data.  Their model, the Sparsity and Delay Embedding-based Forecasting model, or SPADE4, uses machine learning to predict the progression of an epidemic using only limited infection data. SPADE4 was tested on both simulated epidemics and real data from the fifth wave of the Covid-19 pandemic in Canada and successfully predicted the epidemics’ progressions with 95 per cent confidence.  “Covid taught us that we really need to come up with methods ...

Tracking the ol' mutation trail

Tracking the ol mutation trail
2023-08-31
Kyoto, Japan --  From the early stages of cell mutations starting in puberty to their manifestations as breast cancer in later years, the entire process has remained shrouded in mystery. Now, a team of researchers at Kyoto University has revealed the mechanism by which breast cancer is formed in the cells of the mammalian epithelium, whose main function is to secrete milk. According to the team's first analysis, approximately 20 mutations accumulate annually in each epithelial cell until menopause. After menopause, however, the mutation rate significantlydecreases.  "Additionally, our results suggest ...

When the gig is up; gig workers don’t always trust their boss and that might be a good thing

2023-08-31
DURHAM, N.H. — As the so-called ‘gig economy’ continues to grow, so do questions about how this type of non-traditional work compares to full time work arrangements and how these new relationships differ and impact performance and commitment. Researchers from the University of New Hampshire took a closer look at gig workers – which include freelancers, independent contractors and temporary workers – and examined relationships between workers and their managers and found that one trait, trust, could be a double-edged sword. “Millions of workers are now considered gig workers, offering them more flexibility ...

Tracking drivers’ eyes can determine ability to take back control from ‘auto-pilot’ mode

2023-08-31
A team of UCL-led researchers has developed a new method to determine the attention levels of drivers and their readiness to respond to warning signals when using auto-pilot mode. The research, published in Cognitive Research: Principles and Implications, found that people’s attention levels and how engrossed they are in on-screen activities can be detected from their eye movements. The findings suggest a new way to determine the readiness of drivers using auto-pilot mode to respond to real world signals, such as takeover requests from the car. Although fully autonomous driverless cars are not yet available for personal use, cars with a “driverless” auto-pilot ...

Early action to control cardiovascular risk factors preserves brain metabolism

Early action to control cardiovascular risk factors preserves brain metabolism
2023-08-31
Cardiovascular disease and dementia frequently occur together in elderly people. Nevertheless, few longitudinal studies have examined how atherosclerosis and its associated risk factors affect brain health from middle age. Now, a new study by scientists at the Centro Nacional de Investigaciones Cardiovasculares (CNIC) in Madrid provides new data on this relationship; the results confirm the importance of controlling traditional cardiovascular risk factors, such as hypertension, cholesterol, diabetes, smoking, ...

Avoid cannabis during adolescence, pregnancy and while driving, say experts

2023-08-31
Experts recommend avoiding cannabis during adolescence and early adulthood, in people prone to or with mental health disorders, in pregnancy, and before and while driving, based on an in-depth evidence review published by The BMJ today. However, they say cannabidiol (one active compound in cannabis) is effective in people with epilepsy, and cannabis based medicines can help people with multiple sclerosis, chronic pain, inflammatory bowel disease, and in palliative care. Their recommendations are based on an “umbrella review” of 101 meta-analyses on cannabis ...

Call for action over unreliable private online hormone tests

2023-08-31
A large private laboratory is still processing finger prick tests for oestrogen levels, which are sold by private retailers online, despite warnings they are unreliable, reveals an investigation published by The BMJ today.  Journalist Emma Wilkinson reports that Eurofins, a large laboratory based in the UK, is still carrying out finger prick tests for oestradiol despite problems being identified in 2021 and two other laboratories and one online retailer withdrawing the tests over concerns that the results might not always be accurate. Finger prick tests for oestradiol are sold by online retailers for between £50 and £180, depending on what is included in ...

LAST 30 PRESS RELEASES:

QUT scientists create material to turn waste heat into clean power

Major new report sets out how to tackle the ‘profound and lasting impact’ of COVID-19 on cardiovascular health

Cosmic crime scene: White dwarf found devouring Pluto-like icy world

Major report tackles Covid’s cardiovascular crisis head-on

A third of licensed GPs in England not working in NHS general practice

ChatGPT “thought on the fly” when put through Ancient Greek maths puzzle

Engineers uncover why tiny particles form clusters in turbulent air

GLP-1RA drugs dramatically reduce death and cardiovascular risk in psoriasis patients

Psoriasis linked to increased risk of vision-threatening eye disease, study finds

Reprogramming obesity: New drug from Italian biotech aims to treat the underlying causes of obesity

Type 2 diabetes may accelerate development of multiple chronic diseases, particularly in the early stages, UK Biobank study suggests

Resistance training may improve nerve health, slow aging process, study shows

Common and inexpensive medicine halves the risk of recurrence in patients with colorectal cancer

SwRI-built instruments to monitor, provide advanced warning of space weather events

Breakthrough advances sodium-based battery design

New targeted radiation therapy shows near-complete response in rare sarcoma patients

Does physical frailty contribute to dementia?

Soccer headers and brain health: Study finds changes within folds of the brain

Decoding plants’ language of light

UNC Greensboro study finds ticks carrying Lyme disease moving into western NC

New implant restores blood pressure balance after spinal cord injury

New York City's medical specialist advantage may be an illusion, new NYU Tandon research shows

Could a local anesthetic that doesn’t impair motor function be within reach?

1 in 8 Italian cetacean strandings show evidence of fishery interactions, with bottlenose and striped dolphins most commonly affected, according to analysis across four decades of data and more than 5

In the wild, chimpanzees likely ingest the equivalent of several alcoholic drinks every day

Warming of 2°C intensifies Arctic carbon sink but weakens Alpine sink, study finds

Bronze and Iron Age cultures in the Middle East were committed to wine production

Indian adolescents are mostly starting their periods at an earlier age than 25 years ago

Temporary medical centers in Gaza known as "Medical Points" (MPs) treat an average of 117 people daily with only about 7 staff per MP

Rates of alcohol-induced deaths among the general population nearly doubled from 1999 to 2024

[Press-News.org] Better paths yield better AI
Like climbing a mountain via the shortest possible path, improving classification tasks can be achieved by choosing the most influential path to the output, and not just by learning with deeper networks