For a projector at its price point, TNW’s Tristan Greene had a lot of fine things to say, calling the V600 “a fantastic projector.” You can find out for yourself now at over $100 off its regular price, just $144.50 for a limited time when you use Valentine’s Day sale coupon code VDAY2021 during checkout.
The V600 hits a real market sweet spot, offering a true HD 1080p image quality at an extremely affordable price. With a native resolution of 1920×1080 and a contrast ratio of 5000:1, this projector offers up a picture that’s 3 times crisper and more detailed than more traditional 720p projectors on the market.
The greatest strength of the V600 is that that 1080p image also comes powering through admirably in less than perfect lighting conditions.
As Tristan said, “Most budget-priced projectors won’t throw a strong enough picture for gaming in the sun…This one is different.” Whether you’re watching a movie, playing games, or giving a work presentation, the V600 offers remarkable brightness and picture clarity, even in well-lit rooms.
And this projector even maintains much of that image crispness even when cranked out to its maximum 300-inch screen size. That’s a screen measuring 25 feet across, assuming you’ve got a wall big enough to accommodate a picture that large. Plus, the V600 uses an ultra-long lasting lamp bulb, which has been rated to keep going for more than a decade of regular daily use.
An Amazon’s Choice honoree, the V600 also brings the versatility today’s users demand. Featuring a pair of HDMI ports, one with premium audio capabilities, users can connect their laptop, smartphone, or another streaming device easily, while the built-in dual stereo speakers serve up the quality sound to go with your sharp picture.
The Vankyo Performance V600 Projector retails for $249, but in addition to an already healthy discount, shoppers can get an extra 15 percent off for a limited time with theVDAY2021 code. That bringsyour final price down to just $144.50.
TLDR:This subscription to AfterClick Heatmap Analytics offers website heatmapping to track every user’s action on your website in stunningly clear visuals.
This is the age of Big Data. There isn’t an industry on the planet that can’t immediately immerse itself in reams and reams of information about every aspect of that business. But sometimes, all that data isn’t necessarily a good thing.
If you’re a website owner, a quick look at Google Analytics makes the point. Unless you’re a data scientist, there’s a good chance the page after page of breakdowns and statistics attempting to document and explain every visitor to your site and their actions will turn into a numbers equivalent of white noise. It’s just row upon row of figures — and for most, it’s too much.
AfterClick Heatmap Analyticsgets that frustration and answered by creating an analysis interface that non tech-heads could understand quickly and easily. Right now, it’sjust $42.50with the special limited time Valentine’s Day extra discount offer.
A picture really is worth a thousand words. That’s why AfterClick turns all the analytics data about a website’s users into highly visual representations, allowing site owners to understand at a glance what’s happening on their pages.
AfterClick tracks the cursor movements of each website visitor and chronicles their actions. That allows AfterClick users to see each page of their site with an overlay heatmap, showing the areas where the most people are interacting with that page. Whether your user is on your site from a desktop, a smartphone, or a tablet, you’ll know why they’re there — because you can see it in an ultra-clear visual way.
Using AfterClick data, sites can improve their conversion rates by moving call to action items into high traffic areas and know immediately what’s drawing attention and what isn’t on the site. AfterClick also lets you track each user through your site with the Replay feature, which highlights their entire path from page to page through their visit. There’s also Live Users functionality so you can always know exactly how many users are on your site in real-time.
AfterClick is compatible with any content management system or site builder, including WordPress, Shopify, Wix, Weebly, and more.
Right now, a lifetime subscription to AfterClick Heatmap Analytics’ Premium Pro service covering up to 50 websites for tracking up to 250,000 visitors each is over 90 percent off its nearly $3,000 price. On top of that discount, you can also save an additional 15 percent off as part of the limited-time Valentine’s Day sale. Just enter the coupon codeVDAY2021 during checkout to drop your final price down toonly $42.50.
For decades, we’ve been trying to develop artificial intelligence in our own image. And at every step of the way, we’ve managed to create machines that can perform marvelous feats and at the same time make surprisingly dumb mistakes.
After six decades of research and development, aligning AI systems with our goals, intents, and values continues to remain an elusive objective. Every major field of AI seems to solve part of the problem of replicating human intelligence while leaving out holes in critical areas. And these holes become problematic when we apply current AI technology to areas where we expect intelligent agents to act with the rationality and logic we expect from humans.
In his latest book,The Alignment Problem: Machine Learning and Human Values, programmer and researcher Brian Christian discusses the challenges of making sure our AI models capture “our norms and values, understand what we mean or intend, and, above all, do what we want.” This is an issue that has become increasingly urgent in recent years, asmachine learning has found its way into many fields and applications where making wrong decisions can havedisastrous consequences.
As Christian describes: “As machine-learning systems grow not just increasingly pervasive but increasingly powerful, we will find ourselves more and more often in the position of the ‘sorcerer’s apprentice’: we conjure a force, autonomous but totally compliant, give it a set of instructions, then scramble like mad to stop it once we realize our instructions are imprecise or incomplete—lest we get, in some clever, horrible way, precisely what we asked for.”
InThe Alignment Problem,Christian provides a thorough depiction of the current state of artificial intelligence and how we got here. He also discusses what’s missing in different approaches to creating AI.
Here are some key takeaways from the book.
Machine learning: Mapping inputs to outputs
In the earlier decades of AI research,symbolic systemsmade remarkable inroads in solving complicated problems that required logical reasoning. Yet they were terrible at simple tasks that every human learns at a young age, such as detecting objects, people, voices, and sounds. They also didn’t scale well and required a lot of manual effort to create the rules and knowledge that defined their behavior.
But despite their remarkable achievements, machine learning algorithms are at their core complex mathematical functions that map observations to outcomes. Therefore, they’re as good as their data and they start to break as the data they face in the world starts to deviate from examples they’ve seen during training.
InThe Alignment Problem, Christian goes through many examples where machine learning algorithms have caused embarrassing and damaging failures. A popular example is a Google Photos classification algorithm thattagged dark-skinned people as gorillas. The problem was not with the AI algorithm but with the training data. Had Google trained the model on more examples of people with dark skin, it could have avoided the disaster.
“The problem, of course, with a system that can, in theory, learn just about anything from a set of examples is that it finds itself, then, at the mercy of the examples from which it’s taught,” Christian writes.
What’s worse is that machine learning models can’t tell right from wrong and make moral decisions. Whatever problem exists in a machine learning model’s training data will be reflected in the model’s behavior, often in nuanced and inconspicuous ways. For instance, in 2018, Amazon shut down a machine learning tool used in making hiring decisions because its decisions were biased against women. Obviously, none of the AI’s creators wanted the model to select candidates based on their gender. In this case, the model, which was trained on the company’s historical hiring data, reflected problems within Amazon itself.
This is just one of the several cases where a machine learning model has picked up biases that existed in its training data and amplified them in its own unique ways. It is also a warning against trusting machine learning models that are trained on data we blindly collect from our own past behavior.
“Modeling the world as it is is one thing. But as soon as you beginusingthat model, you arechangingthe world, in ways large and small. There is a broad assumption underlying many machine-learning models that the model itself will notchangethe reality it’s modeling. In almost all cases, this is false,” Christian writes. “Indeed, uncareful deployment of these models might produce a feedback loop from which recovery becomes ever more difficult or requires ever greater interventions.”
Human intelligence has a lot to do with gathering data, finding patterns, and turning those patterns into actions. But while we usually try to simplify intelligent decision-making into a small set of inputs and outputs, the challenges of machine learning show that our assumptions about data and machine learning often turn out to be false.
“We need to consider critically… not only where we get our training data but where we get the labels that will function in the system as a stand-in for ground truth. Often the ground truth is not the ground truth,” Christian warns.
Reinforcement learning: maximizing rewards
Another branch of AI that has gained much traction in the past decade isreinforcement learning, a subset of machine learning in which the model is given the rules of a problem space and a reward function. The model is then left to explore the space for itself and find ways to maximize its rewards.
“Reinforcement learning… offers us a powerful, and perhaps even universal, definition of what intelligenceis,” Christian writes. “If intelligence is, as computer scientist John McCarthy famously said, ‘the computational part of the ability to achieve goals in the world,’ then reinforcement learning offers a strikingly general toolbox for doing so. Indeed it is likely that its core principles were stumbled into by evolution time and again—and it is likely that they will form the bedrock of whatever artificial intelligence the twenty-first century has in store.”
Reinforcement learning is behind great scientific achievements such as AI systems that have mastered Atari games, Go, StarCraft 2, and DOTA 2. It has also found many uses in robotics. But each of those achievements also proves that purely pursuing external rewards is not exactly how intelligence works.
For one thing, reinforcement learning models require massive amounts of training cycles to obtain simple results. For this very reason, research in this field has been limited to a few labs that are backed byvery wealthy companies. Reinforcement learning systems are also very rigid. For instance, a reinforcement learning model that plays StarCraft 2 at championship level won’t be able to play another game with similar mechanics. Reinforcement learning agents also tend to get stuck in meaningless loops that maximize a simple reward at the expense of long-term goals. An example is this boat-racing AI that has managed to hack its environment by continuously collecting bonus items without considering the greater goal of winning the race.
“Unplugging the hardwired external rewards may be a necessary part of building truly general AI: because life, unlike an Atari game, emphatically does not come pre-labeled with real-time feedback on how good or bad each of our actions is,” Christian writes. “We have parents and teachers, sure, who can correct our spelling and pronunciation and, occasionally, our behavior. But this hardly covers a fraction of what we do and say and think, and the authorities in our life do not always agree. Moreover, it is one of the central rites of passage of the human condition that we must learn to make these judgments by our own lights and for ourselves.”
Christian also suggests that while reinforcement learning starts with rewards and develops behavior that maximizes those rewards, the reverse is perhaps even more interesting and critical: “Given the behavior, we want from our machines, how do we structure the environment’s rewards to bring that behavior about? How do we get what we want when it iswewho sit in the back of the audience, in the critic’s chair—wewho administer the food pellets, or their digital equivalent?”
Should AI imitate humans?
InThe Alignment Problem, Christian also discusses the implications of developing AI agents that learn through pure imitation of human actions. An example is self-driving cars that learn by observing how humans drive.
Imitation can do wonders, especially in problems where the rules and labels are not clear-cut. But again, imitation paints an incomplete picture of the intelligence puzzle. We humans learn a lot through imitation and rote learning, especially at a young age. But imitation is but one of several mechanisms we use to develop intelligent behavior. As we observe the behavior of others, we also adapt our own version of that behavior that is aligned with our own limits, intents, goals, needs, and values.
“If someone is fundamentally faster or stronger or differently sized than you, or quicker-thinking than you could ever be, mimicking their actions to perfection may still not work,” Christian writes. “Indeed, it may be catastrophic. You’ll do what youwould do if you were them. But you’re not them. And what you do is not whattheywould do if they wereyou.”
In other cases, AI systems use imitation to observe and predict our behavior and try to assist us. But this too presents a challenge. AI systems are not bound by the same constraints and limits as we are, and they often misinterpret our intentions and what’s good for us. Instead of protecting us against our bad habits,they amplify themand they push us toward acquiring the bad habits of others. And they’re becoming pervasive in every aspect of our lives.
“Our digital butlers are watching closely,” Christian writes. “They see our private as well as our public lives, our best and worst selves, without necessarily knowing which is which or making a distinction at all. They, by and large, reside in a kind of uncanny valley of sophistication: able to infer sophisticated models of our desires from our behavior, but unable to be taught, and disinclined to cooperate. They’re thinking hard about what we are going to do next, about how they might make their next commission, but they don’t seem to understand what we want, much less who we hope to become.”
What comes next?
Advances in machine learning show how far we’ve come toward the goal of creating thinking machines. But the challenges of machine learning and the alignment problem also remind us of how much more we have to learn before we can createhuman-level intelligence.
AI scientists and researchers are exploringseveraldifferentwaysto overcome these hurdles and create AI systems that can benefit humanity without causing harm. Until then, we’ll have to tread carefully and beware of how much credit we assign to systems that mimic human intelligence on the surface.
“One of the most dangerous things one can do in machine learning—and otherwise—is to find a model that is reasonably good, declare victory, and henceforth begin to confuse the map with the territory,” Christian warns.
This article was originally published by Ben Dickson onTechTalks, a publication that examines trends in technology, how they affect the way we live and do business, and the problems they solve. But we also discuss the evil side of technology, the darker implications of new tech and what we need to look out for. You can read the original article here.
It is obvious as this Covid-19 pandemic wears on that transport is being repositioned – both in our minds and our lives. Changeable travel restrictions and new virus strains are reshaping our mobility psychology. Despite the rollout of viable vaccines, any ‘return to normal’ for the foreseeable future will involve more work from home, more time spent locally, less socializing outside the household, and fewer trips for work and pleasure. In short, excepting those for whom commuting is not a choice, the future of urban mobility is clearly going somewhere; smaller.
A silver lining in this is that transport planners and journalists are now writing enthusiastically about ’15-minute cities’, where people can live more fully and sustainably, close to home. This solution is exciting and necessary for many reasons. Yet experience shows that turning this vision of sustainable local life into a reality is not so easy – particularly in cities largely designed for cars. The complex machinery of land use and transport policy is hard to rewire.
Meaningful change requires new kinds of cooperation and engagement and, let’s face it, a certain rebalancing of economic, social, and environmental priorities. Now, more than ever, the Sustainable Development Goals (SDGs) provide a roadmap for this recalibration. Yet for many reasons, we see transport and planning professionals struggle to translate the SDGs into specific, community-based change. Despite the ‘disruption’ and sex appeal of new mobility and Mobility as a Service, these tech-driven innovations are not getting to the heart of the issue. So, how do we bridge this gap between aspiration and real sustainable local living?
Understanding everyday life is the key to unlocking sustainable transport
Transport Infrastructure Ireland, one of Ireland’s key national transport agencies, recently issued a brave and useful answer – study women’s travel needs. Why? Because historically women have lived and moved more locally, especially if they have children. This is because gender is the single biggest organizing feature worldwide, and a major factor in travel behavior (seeITF work on Gender in Transport | ITF (itf-oecd.org)). Women do the large majority of local trips for thepurposes of caring for and educating others. They play a profound role in shaping intergenerational mobility choices. Their mobility needs – often centered around local safety, health , and community facilities – and travel with dependants provide us with a roadmap to real and functioning 15-minute cities.
The research studyTravelling in a Woman’s Shoes: Understanding Women’s Travel Needs in Ireland to Inform the Future of Sustainable Transport Policy and Design shows us that understanding car dependency, modal shift, and asmaller scale travel landscape is about understanding everyday life. It illustrates that, when designing and integrating mobility solutions and land use policy, we need to consider that people are situated with families, gendered roles, fears, joys, and risk appetites. Delivering sustainable mobility in the post-pandemic future is about unpacking these ‘situations’. Whilst research was conducted just prior to the global pandemic, the insights have even greater resonance today as people struggle to trust shared transport.
“No one taught me how to cycle on the road – I wish someone would teach me now… I hope my son grows up to be confident enough to cycle on the road.” – Amanda
The everyday stories of women take us beyond the rhetoric about ‘inclusion’ and ‘community,’ introducing voices into the transport discussion that are absent from consultation. These poignant and candid snapshots of everyday life provide the starting point for the next wave of innovation.
“It’s easier to have a car with a baby, you can just put him in his car seat and be done with it” – Nathalie
“My son’s independence is so important to me, it means he’s not attached to me. I know parents who can’t go anywhere [without their autistic child].” – Josie
“Even now in the middle of the day [20 years after being attacked by three men in a car park while 8 months pregnant], there are certain places I wouldn’t walk to for no rational reason at all. There are just alarm bells going off. This has impacted everything – that fear factor is there. I would get into a car at any time of the day and I’d be very conscious. I try not to let it impact my day-to-day life, but it does.” – Siobhan
“It’s the freedom and independence that driving gives you, and the reassurance of knowing that the car is there in case you get a call saying the kids need to get picked up. That’s just too important.” – Karen
“I can walk perfectly, but it’s just fatigue at the end of walking that would get me.” – Lucy
Designing for women: Converting ‘nice-to-have’ to ‘must-haves’
Designing for women is often about things that would be labeled ‘nice-to-haves’ by current design standards (or simply not thought of at all) –but which are the very things that foster a lifetime of confident and loyal sustainable transport use.
It could be more thoughtful mobility near health services; child-size toilet facilities so that kids don’t fall into adult public toilets; safe cycle paths going places that kids need to go; a convivial night-time coffee stand and good lighting at the tram stop, ensuring women feelsafer; services and infrastructure that link up the crèche, local doctor, library, arts precinct, fruit and veg market and perhaps after-school swimming. If these things don’t exist locally, it could be partnering across traditional silos to create them. Reimagining data to tell us about people’ssituations; education for everyone working in the mobility space about the long-lasting trauma that can flow from women’s unsafe mobility experiences – for them and their families; new ways to involve the community in local transport solutions – to name just a few.
Travelling in a Woman’s Shoes is a timely study that holds the clues to designing mobility for resilient local cities, which is in many ways the new priority for all of us. Click here to read Travelling in a Woman’s Shoes: Understanding Women’s Travel Needs in Ireland to Inform the Future of Sustainable Transport Policy and Design.
SHIFT is brought to you by Polestar. It’s time to accelerate the shift to sustainable mobility. That is why Polestar combines electric driving with cutting-edge design and thrilling performance.Find out how.
In recent years, satellites have become smaller, cheaper, and easier to make with commercial off the shelf parts. Some even weigh as little as one gram. This means more people can afford to send them into orbit. Now, satellite operators have started launching mega-constellations – groups of hundreds or even thousands of small satellites working together – into orbit around Earth.
Instead of one large satellite, groups of small satellites can provide coverage of the entire planet at once. Civil, military, and private operators are increasingly using constellations to create global and continuous coverage of the Earth. Constellations can provide a variety of functions, including climate monitoring, disaster management, or digital connectivity, like satellite broadband.
But to provide coverage of the entire planet with small satellites requires a lot of them. On top of this, they have to orbit close to Earth’s surface to reduce interruption of coverage and communication delays. This means they take up an already busy area of space called low Earth orbit, space 100 to 2,000 km above the Earth’s surface.
There are many issues associated with introducing this many satellites into orbit, from the dangers of space junk to obstructing our view of the night sky. But the shift toward mega-constellations is also a challenge for global space governance.
There are almost 3,000 active satellites in orbit around Earth today, and this is set to skyrocket in the coming years. The European Commission, for example, recently announced plans to launch thousands of satellites into orbit around Earth, adding to a growing list of planned mega-constellations launches.
As companies and governments around the world continue to pursue mega-constellations, it is critical that the governance framework is able to support the rise in activity. There are a number of important problems that need to be considered.
Satellites are regulated at the national level and through licensing, guided by the principles of the Outer Space Treaty of 1967. Though the terms constellation or mega-constellation are not found in the treaty, they are considered space objects, like all other satellites.
As procedures and regulations vary from country to country, the challenge is how to govern mega-constellations without creating legal fragmentation. It is imperative that the topic is discussed at the international level.
Yet currently, there is no legally binding definition for a satellite constellation, nor for the newer term mega-constellation. Exactly how many satellites make up a mega-constellation is unknown, and each country could consider the term to mean something different. Clarity at the international level could pave the way for creating guidelines specifically for mega-constellations, which could aid the safe and sustainable use of low Earth orbit.
Most satellites in low Earth orbit are operating between 600 and 800 km above sea level. This is considered a congested area, as there are lots of satellites there already. Small satellites have shorter lifespans than the larger satellites, which typically orbit above low Earth orbit.
However, it can still take up to about 150 years for satellites to be removed, by re-entering the atmosphere and burning up, if they are about 750 km above sea level. Some are removed purposefully, through controlled re-entry, and others are designed to fall in an uncontrolled way. Satellite and mega-constellations operators must consider ways of reducing the debris caused by these satellites above and beyond the usual procedure, in order to maintain sustainable use of low Earth orbit.
Given the number of future mega-constellations currently planned, the space around Earth termed low Earth orbit could easily become a limited resource.
3. Radio spectrum
This is not only true when it comes to physical space, but also radio use. To communicate, satellites use the radio spectrum. With the increase in mega-constellations, there is a danger of operators “warehousing” radio frequencies, stockpiling them before they actually need them.
To prevent this, a United Nations specialized agency for satellite radio spectrum use has recently updated its regulatory framework, dealing with the issue separately from other space regulation. Mega-constellations will be put on a flexible timeline, only being granted use of the frequencies they need at the time.
4. Collision avoidance and tracking
If low Earth orbit becomes overcrowded with satellite and mega-constellations, avoiding collisions will become more difficult. In September 2019, the European Space Agency had to fire the boosters on one of its satellites to get it out of the way of another satellite, otherwise the two would’ve collided.
As the orbit becomes more congested, there may be a need for more collision avoidance maneuvers and better communication between satellite operators.
There are national endeavors, predominately in the United States, for satellite tracking and collision avoidance maneuvering. A system alerts satellite operators to potential collision paths and allows for course corrections where possible.
Hopefully, mega-constellations will be discussed by member states at the UN as soon as they are able to do so. Though work in the committee can be slow and highly political, international guidelines along with national licensing procedures need to add considerations for mega-constellations.
The benefits of constellations and mega-constellations in low Earth orbit for socio-economic and environmental purposes are great. Because of this, it looks likely that the numbers of constellations will increase in the near future. To make sure we avoid problems arising, the rules and definitions surrounding mega-constellations should be made clear, on an international scale.
On Thursday at the World Economic Forum’s annual meeting, he said the dispute had attracted so much attention due to Google’s high levels of transparency:
We are a lot more transparent than most other companies, and so you do see us in the middle of these issues. But I take it as a sign that we allow for debate to happen around this area. And we need to get better as a company, we are committed to doing so, but I look at the state of how we approach this and I am confident about the effort we are putting into it and our commitment to do better here over time.
Gebru chastised his comments in a series of tweets. She called Pichai “a leader in gaslighting” and accused the company’s leadership team of creating “hostile work environments.”
At Davos @sundarpichai gets a question about me & he talks about how Google is a leader in AI ethics & how they “allow debate.” You are a leader in gaslighting. Privileged men like @JeffDean surrounded by other privileged men ordered me to retract a peer reviewed paper… 1
While Pichai’s praise of Google’s transparency sounds more like spin than substance, he’s correct that the dispute has become unusually public. But that’s almost certainly due to growing internal dissent about diversity and ethics at Google.
Gebru has become a powerful symbol of the discontent, which will likely rumble on for many months to come.
Transport concepts like Hyperloop are getting many people hot under the collar. The idea of being able to travel between major cities in minutes rather than hours is certainly an exciting idea. But that’s just it, it’s an idea, and a long long way from reality.
If you need something a little closer to reality to keep your high-speed train/tube travel whistle wet, check out this new magnetic levitation — maglev — train from China.
The prototype unveiled by researchers at Southwest Jiaotong University in Chengdu, China, and is unlike other maglev trains. Rather than using liquid helium superconductors at -269 degrees Celsius, researchers have implemented high-temperature superconductors, which operate at -196 degrees Celsius and are a heck of a lot cheaper to produce and run.
It could bring high-speed intercity train travel to even more people in the country. A commercially viable version of the new maglev train is expected to be released within the next six years.
SHIFT is brought to you by Polestar. It’s time to accelerate the shift to sustainable mobility. That is why Polestar combines electric driving with cutting-edge design and thrilling performance.Find out how.
Have you ever binged on a TV show so much that you fell asleep mid-episode? The last you want is for the show to keep on playing indefinitely, potentially exposing you to spoilers once you wake up and resume the show later (yes, this has happened to me before, and it sucked).
Fret no more: Netflix is testing a timer feature on Android that will let you pause a show after a certain time has elapsed. As reported by the Verge —Netflix confirmed the feature to the publication — the timer function allows you to choose to stop playback after 15, 30, or 45 minutes — or until the end of your current episode. The feature can be accessed from a timer button on the upper right of video playback.
This seems like a useful feature for people who like to fall asleep to a TV show on in the background. While that’s not something I regularly do with Netflix, I do often sleep while listening to audiobooks and find the ability to set a sleep timer to be a godsend.
The test is currently limited to a few users globally on Android, but if it’s popular enough, Netflix will presumably bring it to TVs and other devices. That said, there’s no word on when we might see a wider rollout quite yet.
MIT researchers have invented an adaptive “liquid” neural network that could improve decision-making in self-driving cars and medical diagnosis.
The algorithm adjusts to changes experienced by real-world systems by changing their underlying equations as they receive new data.
“This is a way forward for the future of robot control, natural language processing, video processing — any form of time series data processing,” said Ramin Hasani, the study paper’s lead author. “The potential is really significant.”
Hasani said the system is inspired by a tiny worm — the C. elegans:
It only has 302 neurons in its nervous system yet it can generate unexpectedly complex dynamics.
The code was influenced by the way the C. elegans’ neurons activate and communicate with each other through electrical impulses.
Hasani structed his neural network so that the parameters can change over time based on the results of a nested set of differential equations.
This allows it to continue learning after the training phase, making it more resilient to unexpected situations, like heavy rain covering a camera on a self-driving car.
The liquid network’s small number of highly expressive neurons also makes it easier to interpret its decisions.
“Just [by] changing the representation of a neuron, you can really explore some degrees of complexity you couldn’t explore otherwise,” said Hasani.
In tests, the network performed promisingly in predicting future values in datasets, ranging from atmospheric chemistry to traffic patterns.
Its small size also significantly reduced the computing costs.
Hasani said he now wants to prepare the system for real-world applications:
We have a provably more expressive neural network that is inspired by nature. But this is just the beginning of the process. The obvious question is how do you extend this? We think this kind of network could be a key element of future intelligence systems.
You can read the study paper on the pre-print server arXiv.
When I was six, I thought action figures with Kung-Fu grip were the epitome of technology. Katariya currently has at least six different certifications across AI and data science and what’s sure to be a sky’s-the-limit career in the STEM world.
Leading up to his world record, Kautilya began reading IBM’s course materials to help him understand computer programming and Python language concepts. Due to COVID-19’s impact and the need to stay at home, taking the courses was one way that Kautilya, with his parents’ support, was able to spend time learning new skills.
That kind of makes learning to make sourdough bread seem small by comparison doesn’t it?
Quick take: This is a lovely feel-good story on the surface, but if you peel back a few layers… it’s even better. Winning a Guinness World Record is an amazing feat, but the real story here is how accessible coding/programming has become to the entire world.
Plenty of kids have learned to code at a young age, but Katariya’s the youngest to achieve the certification levels they did. And that just goes to show you that anyone with an internet connection and a curious mind can learn something as complex as coding for artificial intelligence.
File this under: good for the whole community. If you’ve considered learning to code but you’re not sure where to start, here’s a little motivation via IBM’s blog post: