How to Fix Facebook—Before It Fixes Us


new study has overturned a hundred-year-old assumption on what exactly makes a neuron ‘fire’


The human brain contains a little over 80-odd billion neurons, each joining with other cells to create trillions of connections called synapses.

The numbers are mind-boggling, but the way each individual nerve cell contributes to the brain’s functions is still an area of contention. A new study has overturned a hundred-year-old assumption on what exactly makes a neuron ‘fire’, posing new mechanisms behind certain neurological disorders.

A team of physicists from Bar-Ilan University in Israel conducted experiments on rat neurons grown in a culture to determine exactly how a neuron responds to the signals it receives from other cells.

To understand why this is important, we need to go back to 1907 when a French neuroscientist named Louis Lapicque proposed a model to describe how the voltage of a nerve cell’s membrane increases as a current is applied.

Once reaching a certain threshold, the neuron reacts with a spike of activity, after which the membrane’s voltage resets.

What this means is a neuron won’t send a message unless it collects a strong enough signal.

Lapique’s equations weren’t the last word on the matter, not by far. But the basic principle of his integrate-and-fire model has remained relatively unchallenged in subsequent descriptions, today forming the foundation of most neuronal computational schemes.

Image credit: NICHD/Flickr
According to the researchers, the lengthy history of the idea has meant few have bothered to question whether it’s accurate.

“We reached this conclusion using a new experimental setup, but in principle these results could have been discovered using technology that has existed since the 1980s,” says lead researcher Ido Kanter.

“The belief that has been rooted in the scientific world for 100 years resulted in this delay of several decades.”

The experiments approached the question from two angles – one exploring the nature of the activity spike based on exactly where the current was applied to a neuron, the other looking at the effect multiple inputs had on a nerve’s firing.

Their results suggest the direction of a received signal can make all the difference in how a neuron responds.

A weak signal from the left arriving with a weak signal from the right won’t combine to build a voltage that kicks off a spike of activity. But a single strong signal from a particular direction can result in a message.

This potentially new way of describing what’s known as spatial summation could lead to a novel method of categorising neurons, one that sorts them based on how they compute incoming signals or how fine their resolution is, based on a particular direction.

Better yet, it could even lead to discoveries that explain certain neurological disorders.

It’s important not to throw out a century of wisdom on the topic on the back of a single study. The researchers also admit they’ve only looked at a type of nerve cell called pyramidal neurons, leaving plenty of room for future experiments.

But fine-tuning our understanding of how individual units combine to produce complex behaviours could spread into other areas of research. With neural networks inspiring future computational technology, identifying any new talents in brain cells could have some rather interesting applications.

This research was published in Scientific Reports.

The Wild Week in AI – Andrew Ng’s new Manufacturing AI company; Google’s China Lab; Deep Learning Trends Tutorial; and more | Revue

— except below —

The Wild Week in AI – Andrew Ng’s new Manufacturing AI company; Google’s China Lab; Deep Learning Trends Tutorial; and more

The Wild Week in AI
The Wild Week in AI
December 18 · Issue #72 · View online
The Wild Week in AI is a weekly AI & Deep Learning newsletter curated by @dennybritz.
If you enjoy the newsletter, please consider sharing it on Twitter, Facebook, etc! Really appreciate the support 🙂
This Week’s Sponsor: Butterfly Network
Butterfly has built a new hand-held Ultrasound device that will revolutionize health care: the Butterfly iQ. This Ultrasound device fits in your pocket, connects to your smart-phone and stores medical data securely in the cloud.
Butterfly’s machine learning team works on building intelligence into the device to help clinicians make life-saving decisions. Butterfly is looking for researchers interested in continuing to develop and publish new machine learning algorithms while also having a direct and immense, real-world impact. Find out more at
Andrew Ng launches AI + Manufacturing Startup
Founded by famous professor Andrew Ng, is a new Artificial Intelligence company focused on the manufacturing industry. At this point, it is still unclear what kind of products is working on.
Google opens AI Center in Beijing, China
The Google AI China Center will have a small group of researchers supported by several hundred China-based engineers. “It will be a small team focused on advancing basic AI research in publications, academic conferences, and knowledge exchange,“ said Fei-Fei Li, the chief scientist at Google’s cloud unit who will lead the Beijing research center.
AlphaGo Teach: Discover new and creative ways of playing Go
This tool provides analysis of 6,000 of the most popular opening sequences from the recent history of Go, using data from 231,000 human games and 75 games AlphaGo played against human players.
AI-Assisted Fake Adult Videos
Someone used a Machine Learning algorithm to paste the face of ‘Wonder Woman’ star Gal Gadot onto an adult video. It’s not going to fool anyone who looks closely. Sometimes the face doesn’t track correctly and there’s an uncanny valley effect at play, but at a glance, it seems believable.
Posts, Articles, Tutorials
Deep Learning: Practice and Trends (NIPS 2017 Tutorial)
An excellent tutorial on the building blocks of today’s Deep Learning systems. The tutorial covers Convolutional Models, Autoregressive Models, Domain Alignment, Meta Learning, Graph Networks, and more.
Deep Learning for NLP – Advancements and Trends in 2017
A good summary of Deep Learning advancements for NLP in 2017. This post covers, pre-trained word embeddings, the sentiment neuron, SemEval 2017 results, abstractive summarization systems, unsupervised Machine Translation, and more.
Introduction to Gaussian Processes
Gaussian processes may not be at the center of current machine learning hype but are still used at the forefront of research – they were recently seen automatically tuning the MCTS hyperparameters for AlphaGo Zero for instance.
Training Sequence Models with Attention
Several practical tips for training sequence-to-sequence models with attention, such as those used in Machine Translation, or text summarization.
Code, Projects & Data
MAgent Platform for Many-agent Reinforcement Learning
MAgent is a research platform for many-agent reinforcement learning. Unlike previous research platforms that focus on reinforcement learning research with a single agent or few agents, MAgent aims at supporting reinforcement learning research that scales up from hundreds to millions of agents.
Visual to Sound: Generating Natural Sound for Videos in the Wild
In this paper and project, the authors pose generate sound given visual input and apply learning-based methods to generate raw waveform samples given input video frames.
Exploring the ChestXray14 dataset: Problems
A detailed analysis of the ChestXray14 dataset, and why it may not be fit for training medical AI systems to do diagnostic work. Such analyses of real-world datasets are extremely important, and I hope to see more of them in the future.
Highlighted Research Papers
Libratus AI for heads-up no-limit poker (Science)
The authors present Libratus, an AI that, in a 120,000-hand competition, defeated four top human specialist professionals in heads-up no-limit Texas hold’em, the leading benchmark and long-standing challenge problem in imperfect-information game solving. The game-theoretic approach uses application-independent techniques: an algorithm for computing a blueprint for the overall strategy, an algorithm that fleshes out the details of the strategy for subgames that are reached during play, and a self-improver algorithm that fixes potential weaknesses that opponents have identified in the blueprint strategy.
[1712.03351] Peephole: Predicting Network Performance Before Training
An approach to predict the performance of a network before training, based on its architecture. The authors develop a way to encode individual layers into vectors and bring them together to form an integrated description via LSTM. Taking advantage of the recurrent network’s expressive power, this method can reliably predict the performances of various network architectures.
[1712.04741] Mathematics of Deep Learning
Recently there has been a dramatic increase in the performance of recognition systems due to the introduction of deep architectures for representation learning and classification. However, the mathematical reasons for this success remain elusive. This tutorial will review recent work that aims to provide a mathematical justification for several properties of deep networks, such as global optimality, geometric stability, and invariance of the learned representations.
Did you enjoy this issue?
Thumbs up 1ae5a7bdfcd3220e2b376aa0c1607bc5edaba758e5dd83b482d03965219a220b Thumbs down e13779fa29e2935b47488fb8f82977fedcf689a0cc0cc3c19fa3c6bb14d1493b

Carefully curated by The Wild Week in AI with Revue.
If you were forwarded this newsletter and you like it, you can subscribe here.
If you don’t want these updates anymore, please unsubscribe here.

Share to Twitter Share to Facebook
Subscribe to my list

Your email address…
Subscribe now
This thingy has to be empti pwieshh

Proudly crafted with

Noise Pollution health crisis looms

— excerpt below —

Cities don’t just get filthy in the visual sense — urban spaces can often be a raging mess of sound. The blaring car horns, the police sirens screaming through the usual waves of traffic, the clatter of passing trains, the constant cacophony of voices rising and falling as a strange melody — this deluge of noise can be devastating to the human psyche.

“I think, as a society, we are becoming much more aware of the noise around us,” Scott Sommerfeldt, an engineer at Brigham Young University who specializes in acoustic noise mitigation, tells Futurism. “Excessive noise has detrimental health effects on us, and we’re finally realizing how those effects add up.”

By 2100, 84 percent of the world’s estimated 10.8 billion people will likely live in cities. That means noise pollution will bloom in those areas and beyond, in surrounding suburbs and rural spaces that were typically safe havens from the clamor of the city. Urban sprawl will get worse, encompassing quieter areas into the thick sounds of city life.

It’s impossible to overstate how much noise pollution can wreak havoc on human health and safety. High noise levels can exacerbate hypertension, cause insomnia or sleep disturbances, result in hearing loss, and worsen a plethora of other medical conditions. All of these problems can aggravate other health issues by inducing higher levels of stress, which can cascade into worsened immune systems, heart problems, increased anxiety and depression — the list just goes on and on.

Humans have been suffering because of excess noise for a long time, and scientists have known about it at least for several decades. The government first acknowledged that noise was a problem when it passed the Noise Pollution and Abatement Act of 1972. “There was already a real awareness that noise was harmful to health, and the Act was a formal acknowledgment by the government,” says Arline Bronzaft, a psychologist and authority on noise pollution who currently serves as chairman of the noise committee of GrowNYC. “Although there wasn’t enough information to point to how noise affects us in every which way, there was enough for us to move forward with back in 1972. And today, there’s enough literature for everyone to agree noise pollution is a public health hazard.”

Humans aren’t the only species suffering from the racket. Urban and suburban noise can easily escape from peopled areas into nature, causing problems for animal populations, especially those that live close to highways or busy harbors. Noise pollution can make it harder for animals to avoid predators (or for predators to find and catch prey), reproduce, and live healthy lives.

Birds and marine mammals, for instance, rely on specific vocalizations to attract potential mates. Noise pollution can disrupt those behaviors. Bats rely on echolocation to move around and find food but have an impaired ability to forage when ambient noise levels are high. Oysters clam up during stress; exposure to loud noises along the coast has caused them to close their shells more frequently and for longer periods, preventing them from ingesting enough food. As filter feeders, oysters play a critical role in keeping the ocean’s water quality at a healthy, sustainable level for all marine life. They can’t do that if they’re all clammed up, which could cause the entire habitat to deteriorate.

Where is most of this noise coming from? Traffic is the biggest contributor to noise pollution. A diesel truck at 50 feet away, for instance, generates up to 90 decibels of noise. Generally speaking, prolonged exposure to anything over 85 decibelsputs someone at risk for temporary or permanent hearing loss.

But noise pollution is more than just automobiles. Increased development — in the big city or in a quaint suburb — means construction sites, where heavy machinery creates a fitful, ugly noise that can echo into the placid surrounding areas. People living close to train tracks or airports are bludgeoned with noise (and usually accompanying vibrations and shakes) at all hours. Air traffic can be a major headache. All these factors are exacerbated by city planning and community zoning, which fail to mitigate noise for residents.

So all things considered, what can be done to keep future society from turning into a deaf, noise-addled dystopia? Bronzaft says there’s more awareness about noise pollution these days among the general public than there was in 1972, but we need more than just awareness. As cities swell up to more extreme sizes, without a responsible combination of technological innovations and more radical policy measures, the problem is bound to get worse.

One approach is to think big. Roadway or railway barriers can buffer communities from the noise of passing cars and trains. European engineers, in particular, have developed novel building materials and barrier designs (such as curved structures, and steel or plexiglass) to envelop highway noise, leaving the outside with little more than a quiet hum.

Barriers to highway noise. Image credit: Reider Group

Hybrids and electric cars generate quieter noise levels than engines that guzzle gasoline and diesel. In mass numbers, even these greener vehicles cause a lot of noise, mostly from tire and road friction and wind passage. But they do mean individual cars can sidle through quiet streets without causing much of ruckus (in fact, electric vehicles are sometimes too quiet for comfort — the government requires that these cars make at least some minimal threshold of noise at low speeds in order to give other drivers and bystanders a heads up).

Within cities themselves, developers who are keen to rent quieter office and residential space for a premium are more inclined now to fit new buildings with thicker glass facades or insulation materials that limit noise. Interior designers are outfitting buildings with soundproofing materials, such as acoustic ceiling tiles. Boilers and generators are typically secluded into basement rooms that will not disturb individuals on other floors.

Policy initiatives have also made changes for the better in the past. Bronzaft coauthored a landmark 1975 study that demonstrated the negative impact of noise on student performance in schools. “Here, I do one study on the effects of noise in schools, and the [New York City] Transit Authority quiets the track adjacent to the school, and the Board of Education puts acoustic ceilings in the classrooms,” Bronzaft says. After those changes were made, researchers conducted a follow-up study and found a significant improvement in student reading comprehension. More recently, in 2007, Bronzaft was a major force in helping revise New York City’s noise code, which focused largely on limiting construction site noise and even recommended what types of machinery and tools site workers ought to be using to protect the neighbors.

Bronzaft also points out that community groups such as Noise Pollution Clearinghouse, US-Citizens Aviation Watch, and Noise Free Americahave had huge success in mandating changes in aircraft routes to protect people on the ground from jet engine roars, loud traffic, and construction. Bronzaft has been involved in efforts in places like New Orleans, which is renowned for its live music, to draft new noise ordinance revisions. She also helped stop a noisy motocross racetrack from being built in Cheyenne, Wyoming. In her view, these instances are evidence that “people can do something. An ordinary citizen can initiate a positive change.”

Bronzaft is especially bullish on educating children and young adults into taking local action, especially since noise control has been effectively ceded by the U.S. Environmental Protection Agency (EPA) to state and local governments. She lauds the NYC Department of Environmental Protection’s online curriculum on sound and noise, “to teach children about the beauty of sound and the dangers of noise,” as a great example of how to embed noise pollution awareness early. “You should always start with children,” she says.

Given how ineffable noise can be, many people can’t afford to wait for policy changes or large infrastructure projects to help them find some quiet. Luckily, personal technology solutions are becoming more popular. Sommerfeldt is an expert in a field called Active Noise Control (ANC) — the method of using noise to cancel out noise. “This is generally a low-frequency solution,” he says. A popular example is noise-canceling headphones, which protect the user’s ears from outside sound by generating their own white noise. Some cars are also beginning to use ANC solutions to block out external noise (especially from the engine) for passengers inside a vehicle. “It’s relatively easy to implement for cars — up to a certain point,” says Sommerfeldt. “It’s more challenging to make it quiet over a large region or the region completely inside an automobile or plane.”

Critics might decry all of this spending on an issue that doesn’t seem to be immediately hazardous, but Bronzaft emphasizes that the problems caused by noise compound over time if not addressed. Solving noise pollution is a preemptive measure that can forestall bigger physiological and learning issues people may later develop. “Health and student education cost a lot of money in this country,” says Bronzaft. “It’s cheaper in the long term to quiet things down.”

Ultimately, a game-changing factor might simply be accruing more data on the subject so that policy-makers and engineers can put it to use. Some researchers are trying to leverage new technologies to solve that obstacle. A team of scientists at New York University’s Music and Audio Research Lab (MARL) recently launched the Sounds of New York City (SONYC) initiative, which seeks to distribute compact sensors throughout NYC to generate an acoustic “map” of the city to better understand how sounds get dispersed in the city, and what areas are louder than others. As the city that never sleeps with a din to fit, NYC is a pretty obvious testbed for such a project. If SONYC is successful, its data could be combined to health and safety studies to demonstrate more about how debilitating noise pollution is to our lives, and to wildlife populations in nearby regions.

It’s impossible to conceive of a quiet future for the planet — surely, things will only get noisier. True quiet will be increasingly hard to find. But new innovations could go a long way in helping the average person better navigate through the hullabaloo of a clamorous world. There’s little reason to believe it’s inevitable we’ll all go deaf by 55.