Postagem em destaque

Antologia: Miríade, Distopia, Utopia (2004-2024) -

     Antologia : Miríade, Distopia, Utopia  (2004-2024); @vanres1974; #antologia;  {11dez24 qua 20:40-20:50}      Anthology: Myriad, Dystopi...

Prof. Dr. Vander Resende, Doutorado em Lit Bras, pela UFMG; Mestre em Teorias Lit e Crít Cul, UFSJ

quinta-feira, 2 de abril de 2020

Learning difficulties and poorly connection hubs

020420  02mar2020 
children's brains were organised around hubs,
     like an efficient
         traffic system or
          social network.
Those who had well-connected brain hubs
     had either very specific cognitive difficulties,
         such as poor listening skills,
     or had no cognitive difficulties at all.
By contrast, those with poorly connected hubs
     had widespread and severe cognitive problems.

...

This work suggests, he adds, that interventions should be less reliant on diagnostic labels.
"It's better to look at their
     areas of cognitive difficulties and
     how these can be supported,
         for example using specific interventions to improve
             listening skills or
             language competencies,
     or at interventions that
         would be good for the whole class,
             like how to reduce working memory demands
                 during learning." (???)

The findings also may explain why drugs treatments have not been effective for developmental disorders, the researchers say.
Drugs tend to target
     specific types of nerve cells
     but would have little impact on
         a hub-based organisation.
 

How deep learning works?


020420  02mar2020 


if a deep neural network is trained to
     identify objects in photos,
         it will employ different neurons to recognize a photo of a cat than it will to recognize a school bus.

"You don't need to train all the neurons on every case," Medini said.
Deep learning networks were inspired by biology, and their central feature, artificial neurons, are small pieces of computer code that can learn to perform a specific task.

A deep learning network can contain
     millions or even billions of 
     and working together they can
         learn to make
             human-level,
             expert decisions
                 simply by studying large amounts of data.

For example, if a deep neural network is trained to
     identify objects in photos,
         it will employ different neurons to recognize a photo of a cat than it will to recognize a school bus.

"You don't need to train all the neurons on every case," Medini said.
"We thought, 'If we only want to pick the neurons that are relevant, then it's a search problem.'
So, algorithmically, the idea was to use locality-sensitive hashing to get away from matrix multiplication."
Hashing is a data-indexing method invented for internet search in the 1990s. It uses numerical methods to encode large amounts of information, like entire webpages or chapters of a book, as a string of digits called a hash. Hash tables are lists of hashes that can be searched very quickly.
...
Shrivastava said SLIDE hasn't yet come close to reaching its potential.
"We've just scratched the surface," he said. "There's a lot we can still do to optimize. We have not used vectorization, for example, or built-in accelerators in the CPU, like Intel Deep Learning Boost. There are a lot of other tricks we could still use to make this even faster."
Shrivastava said SLIDE is important because it shows
there are other ways to implement .

Irregular Sleep and cardiovascular disease

020420  02mar2020
Older adults with irregular sleep patterns—meaning they have no
     regular bedtime and
     wakeup schedule,
or they get different amounts of sleep each night—
     are nearly twice as likely to develop
cardiovascular disease
as those with more regular sleep patterns,
according to a new study funded in part by the National Heart, Lung, and Blood Institute (NHLBI), part of the National Institutes of Health.

The five-year study suggests that an irregular sleep pattern may be a novel and independent risk factor for cardiovascular disease (CVD), and that

maintaining regular sleep patterns
     could help prevent heart disease just as
physical activity,
a healthy diet,
and other lifestyle measures do, the researchers said.

"Research has linked irregular sleep schedules to a constellation of disease-causing abnormalities in body function, including changes in
blood sugar and
inflammation,"
said Michael Twery, Ph.D., director of the NHLBI's National Center on Sleep Disorders Research.
"This study is Important because it is among the largest of its kind, and it specifically associates these irregular sleep patterns with an increased risk for cardiovascular disease."
...
The researchers said they are still unclear about the exact biological mechanisms behind the sleep irregularity and CVD link, but they suspect that multiple factors, including
harmful disturbances in the body's circadian rhythm
     —the 24-hour internal body clock which controls the sleep-wake cycle—
may be in play.

Recent studies by the same researchers linked
irregular sleep patterns to
     harmful metabolic changes associated with
     obesity,
     diabetes, and
     high cholesterol,
         and they suspect similar processes might also influence 
     CVD risk.
...
In future studies, the researchers said they will look for
     blood biomarkers
         that may help explain the apparent link.

Larger studies with longer follow-up will also be important to confirm these findings.

A better understanding of the root causes could lead to new interventions that could help people better regulate their sleep patterns, such as
     wearing devices or
     using specialized mobile apps
         designed to reduce irregular sleep,
             the researchers said.
Medicinal interventions also may be possible, they noted.

Running and a long life

02apr20 04mar20
participation in running, regardless of its dose, would probably lead to
substantial improvements in
     population health and    
     longevity.
Any amount of running, even just once a week, is better than no running,
but higher doses of running may not necessarily be associated with greater mortality benefits.

Fourteen studies from six prospective cohorts with a pooled sample of 232 149 participants were included. In total, 25 951 deaths were recorded during 5.5–35 year follow-ups. Our meta-analysis showed that

running participation is associated with
27%, 30% and 23% lower risk of
     all-cause,
     cardiovascular and
     cancer mortality, respectively,
                         compared with no running.

A meta-regression analysis showed no significant dose–response trends for weekly frequency, weekly duration, pace and the total volume of running.

Five tips for teaching online

02ap20 24mar2020
Don’t convert entire lecture to video;
Don’t rely on live video;
Pay attention to engagement and feedback;
Check in with students often;
Identify struggling students and support them

Don’t convert your entire lecture to video. 
“Strategically reduce your goals” . Instructors need to identify a few specific things that they want their students to learn, and focus on those.

Don’t rely on live video. 


Invite student engagement and feedback.
The best online instructors, Reich and Lee agree, set up their courses so that students can pursue self-paced enquiry — exploring the topic under their own initiative. For example, you can give students a range of links for further reading.
Asking students what they hope to get out of the online course, and
how you can best serve them,
       offers instructors ideas for teaching and gives students ownership of the process, he says.


Check in with students often. 
Consider interactive elements such as short quizzes.

Students might also miss on-campus social interaction, so it helps to engage them with opportunities to talk to one another during a live session, says Lee. She groups students into teams of five so that they can support one another. If students have a question about content, they ask their group first, and come to her only if they still can’t get a satisfactory answer. “Make sure students support each other. Don’t try to do everything yourself,” she says.

Identify and support struggling students. 
Reich says that the most successful virtual teachers conduct frequent assessments, and check in by phone, text or e-mail with each student — most often with those who are struggling.
To identify those students, instructors can ask whether class members have adequate Wi-Fi and access to devices, and how concerned they are about the transition to online learning. 

With technological help from colleagues at NYU Shanghai, [Leonardo Rolla] developed a strategy for teaching remotely from the other side of the world.

Each day, using a program called Voice-Thread,
he records several short videos of himself explaining maths concepts,
adding up to 15–30 minutes collectively.

During their day, the students
watch the videos on a website and, in turn,
insert videos they make of their assigned theorem proofs, for example, or
a question,
a comment or
a critique of a classmate’s proof.
Together, Rolla and his students produce an interactive, if asynchronous, class recording.

Working together

Rolla has one crucial tip:
seek constant feedback from students.
“I am the director of this movie,” he says, “but we are all in this together.”
He asks his students precise questions to demonstrate
what they have just learnt and
how each concept builds on their existing base of knowledge.

He also asks for feedback to improve the course.

When students asked for more concrete examples of complex, abstract theorems to make sure they understood the concepts, he obliged.

“The biggest risk is that you become a talking head
     explaining things that students are not following,” he says,
     “and they give up and
     just pretend.”
His video-based approach has earned high praise from students and colleagues.
“The hardest part is that it takes a lot of time,” says Rolla, “at least three times as much work as a traditional lecture, and that’s once you’re familiar with the tools.” He estimates that since February, he’s been spending 35 hours weekly preparing his online teaching content.

https://www.nature.com/articles/d41586-020-00896-7