Generative Adversarial Networks (GANs) are a fascinating class of machine learning models used to generate new data that resembles the training data. They were first introduced by Ian Goodfellow and his colleagues in 2014. GANs are particularly popular in the field of image generation but have applications in other areas as well.

Here’s how GANs generally work:

1. Architecture

A GAN consists of two main parts:

• Generator: This component generates new data instances.
• Discriminator: This component evaluates them. It tries to distinguish between real data (from the training dataset) and fake data (created by the generator).

2. Training Process

The training of a GAN involves the following steps:

• The generator takes a random noise vector (random input) and transforms it into a data instance.
• The discriminator receives either a generated data instance or a real data instance and must determine if it is real or fake.

3. Adversarial Relationship

The core idea behind GANs is based on a game-theoretical scenario where the generator and the discriminator are in a constant battle. The generator aims to produce data that is indistinguishable from genuine data, tricking the discriminator. The discriminator, on the other hand, learns to become better at distinguishing fake data from real data. This adversarial process leads to improvements in both models:

• Generator’s Goal: Fool the discriminator by generating realistic data.
• Discriminator’s Goal: Accurately distinguish between real and generated data.

4. Loss Functions

Each component has its loss function that needs to be optimized:

• Discriminator Loss: This aims to correctly classify real data as real and generated data as fake.
• Generator Loss: This encourages the generator to produce data that the discriminator will classify as real.

5. Backpropagation and Optimization

Both the generator and the discriminator are typically neural networks, and they are trained using backpropagation. They are trained simultaneously with the discriminator adjusting its weights to get better at telling real from fake, and the generator adjusting its weights to generate increasingly realistic data.

6. Convergence

The training process is ideally stopped when the generator produces data that the discriminator judges as real about half the time, meaning the discriminator is essentially guessing, unable to distinguish real from fake effectively.

Example Use Cases:

• Image Generation: GANs can generate realistic images that look like they could belong to the training set.
• Super Resolution: Enhancing the resolution of images.
• Style Transfer: Applying the style of one image to the content of another.
• Data Augmentation: Creating new training data for machine learning models.

GANs have been revolutionary due to their ability to generate high-quality, realistic outputs, making them a powerful tool in the AI toolkit. However, training GANs can be challenging due to issues like mode collapse (where the generator produces a limited diversity of samples) and non-convergence.

The post How do generative models like GANs (Generative Adversarial Networks) work? appeared first on Artificial Intelligence.

The possibilities opened up to us by the rise of the Internet of Things (IoT) is a beautiful thing. However, not enough attention is being paid to the software that goes into the things of IoT. This can be a daunting challenge, since, unlike centralized IT infrastructure, there are, by one estimate, at least 30 billion IoT devices now in the world, and every second, 127 new IoT devices are connected to the internet.

Many of these devices aren’t dumb. They are increasingly growing sophisticated and intelligent in their own right, housing significant amounts of local code. The catch is that means a lot of software that needs tending. Gartner estimates that right now, 10 percent of enterprise-generated data is created and processed at the edge, and within five years, that figure will reach 75 percent.

For sensors inside a refrigerator or washing machine, software issues mean inconvenience. Inside automobiles or vehicles, it means trouble. For software running medical devices, it could mean life or death.

“Code rot” is one source of potential trouble for these devices. There’s nothing new about code rot, it’s a scourge that has been with us for some time. It happens when the environment surrounding software changes, when software degrades, or as technical debt accumulates as software is loaded down with enhancements or updates.

It can bog down even the most well-designed enterprise systems. However, as increasingly sophisticated code gets deployed at the edges, more attention needs to be paid to IoT devices and highly distributed systems, especially those with critical functions. Jeremy Vaughan, founder of CEO of TauruSeer, recently sounded the alarm on the code running medical edge environments.

Vaughan was spurred into action when the continuous glucose monitor (CGM) function on a mobile app used by his daughter, who has had Type-1 Diabetes her entire life, failed. “Features were disappearing, critical alerts weren’t working, and notifications just stopped,” he stated. As a result, his nine-year-old daughter, who relied on the CGM alerts, had to rely on their own instincts.

The apps, which Vaughan had downloaded in 2016, were “completely useless” by the end of 2018. “The Vaughans felt alone, but suspected they weren’t. They took to the reviews on Google Play and Apple App store and discovered hundreds of patients and caregivers complaining about similar issues.”

Code rot isn’t the only issue lurking in medical device software. A recent study out of Stanford University finds the training data used for the AI algorithms in medical devices are only based on a small sample of patients. Most algorithms, 71 percent, are trained on datasets from patients in only three geographic areas — California, Massachusetts and New York — “and that the majority of states have no represented patients whatsoever.” While the Stanford research didn’t expose bad outcomes from AI trained on the geographies, but raised questions about the validity of the algorithms for patients in other areas.

“We need to understand the impact of these biases and whether considerable investments should be made to remove them,” says Russ Altman, associate director of the Stanford Institute for Human-Centered Artificial Intelligence. “Geography correlates to a zillion things relative to health. “It correlates to lifestyle and what you eat and the diet you are exposed to; it can correlate to weather exposure and other exposures depending on if you live in an area with fracking or high EPA levels of toxic chemicals – all of that is correlated with geography.”

The Stanford study urges the employment of larger and more diverse datasets for the development of AI algorithms that go into devices. However, the researchers caution, obtaining large datasets is an expensive process. “The public also should be skeptical when medical AI systems are developed from narrow training datasets. And regulators must scrutinize the training methods for these new machine learning systems,” they urge.

In terms of the viability of the software itself, Vaughan cites technical debt accumulated with within medical device and app software that can severely reduce their accuracy and efficacy. “After two years, we blindly trusted that the [glucose monitoring] app had been rebuilt,” he relates. “Unfortunately, the only improvements were quick fixes and patchwork. Technical debt wasn’t addressed. We validated errors on all devices and still found reviews sharing similar stories.” He urges transparency on the components within these devices and apps, including following US Food and Drug Administration guidelines that call for a “Cybersecurity Bill of Materials (CBOM)” that lists out “commercial, open source, and off-the-shelf software and hardware components that are or could become susceptible to vulnerabilities.”

More and more computing and software development is moving to the edge. The challenge is applying the principles of agile development, software lifecycle management and quality control learned over the years in the data center to the edges, and applying automation on a vaster scale to keep billions of devices current.

The post When ‘code rot’ becomes a matter of life or death, especially in the Internet of Things appeared first on Artificial Intelligence.

Is the world too chaotic for any technology to control? Is technology revealing that things are even more chaotic and uncontrollable than first thought? Artificial intelligence, machine learning and related technologies may be  underscoring a realization Albert Einstein had many decades ago: “The more I learn, the more I realize how much I don’t know.”

When it comes to employing the latest analytics in enterprises and beyond, even the best technology — predictive algorithms, artificial intelligence — can’t explain, and even reveal, the complexity and interactions that shape events and trends.  That’s the word from Harvard University’s David Weinberger who explains, in his latest book, how AI, big data, science and the internet are all revealing a fundamental truth: things are more complex and unpredictable than we’ve allowed ourselves to see.

“Our unstated contract with the universe has been if we work hard enough and think clearly enough, the universe will yield its secrets, for the universe is knowable, and thus at least somewhat pliable to our will,” Weinberger writes in Everyday Chaos: Technology, Complexity, and How We’re Thriving in a New World of Possibility. “But now that our tools, especially machine learning and the internet, are bringing home to us the immensity of the data and information around us, we’re beginning to accept that the true complexity of the world far outstrips the laws and models we devise to explain it.”

The irony is the systems we are creating to make some sense of the world — such as machine learning and deep learning — are only creating more mystery. To illustrate this point, Weinberger points to Deep Patient, an AI program developed by Mount Sinai Hospital in New York in 2015. Deep Patient was fed medical records of 700,000 patients as jumbled data, with no framework to organize it or any instructions as to how the data should be used. Yet, even with just three incomplete pieces to analyze, Deep Patient is able to diagnose the likelihood of patients developing some diseases more accurately than doctors.

There’s only one catch to Deep Patient’s success– no one knows why or how it comes to its conclusions. “The number and complexity of contextual variables mean that Deep Patient simply cannot always explain its diagnoses as a conceptual model that its human keepers can understand.,” Weinberger says.

The bottom line, Weinberger says, is that success with AI and automation stems from accepting and leveraging results delivered, and not trying to decipher the explanations behind the data fed into these systems. If A/B testing shows text on a website results in more traffic than a photo placement, go with it.

He outlines four rules of the road that should guide our expectations when it comes to machine learning and deep learning applications:

• Prioritize the workflow of AI systems, but leave it to AI to determine how results are produced.  Today’s deep learning models “are not created by humans, at least not directly. Humans choose the data and feed it in, humans head the system toward a goal, and humans can intercede to tune the weights and outcomes. But humans do not necessarily tell the machine what features to look for. Because the models deep learning may come up with are not based on the models we have constructed for ourselves, they can be opaque to us.”
• Throw away the old models that shaped expectations. “Deep learning models are not generated premised on simplified principles, and there’s no reason to think they are always gong to produce them.”
• We shouldn’t be expected to understand what underlies AI decisions. “Deep learning systems do not have to simplify the world to what humans can understand,” Weinberger says. “Our old, simplified models were nothing more than the rough guess of a couple of pounds of brains trying to understand a realm in which everything is connected to, and influenced by, everything.”
• Ultimately, it’s about the data. “Everything connected to everything means that machine learning’s model can constantly change. Changes in machine learning models can occur simply by retraining them on new data. indeed, some systems learn continuously.”

A question that comes out of this, of course, is whether to trust AI output when bias needs to be questioned or analyzed. While Weinberger’s book does not tackle the issues of human bias being baked into AI algorithms head-on — that is the subject of entire books in and of themselves — he points out that human biases find their way into results, which illustrate our own failings.  People seek explicability “to prevent AI from making our biased culture and systems even worse than they were before AI. Keeping AI from repeating, amplifying and enforcing existing prejudices is a huge and hugely important challenge.”

No matter how automated our decision-making becomes, critical thinking — human critical thinking — is still needed to run businesses and institutions. Humans need to be able to override or question the output of AI systems, especially if the process is opaque.  This is a critical skill that needs to part of every job, every training program and every course curriculum from now on.

The post Four Rules To Guide Expectations Of Artificial Intelligence appeared first on Artificial Intelligence.

Doomsday theorists say Artificial Intelligence and the machines that use it will destroy mankind. While that may be a stretch, there is some merit in the argument that AI will surely challenge human thinking and behaviour – to what outcome, that remains to be seen.

Take the example of a ride-hailing app – Uber, Ola or any other. You, the passenger, and the driver may know the route to a destination, but both are compelled to follow the one calculated by the AI engine or else the driver is penalised. Human intelligence such as experience with traffic patterns, temporary blockages due to repairs and construction, or even shorter routes off the main road are not taken into consideration.

Seen at a macro level, would this create a loss of identity where human intelligence and experience of both the driver and the customer is overridden by a machine? In simple terms, here, a human has been displaced by AI. We may not accept it, but this is, in essence, a machine planning what is best for a human.

Man’s inventions have sought to reduce human effort and this is apparent now more than ever. Every time electronic mediums evolve, they tend to make services obsolete, and now, they are making human decision-making obsolete too.

What is the basis that machines take decisions on? The answer here is data – hundreds and thousands of gigabytes of data is collected across the world – its collected when you walk into a store and buy a toothpaste, it is collected when you pay for a ticket online, its collected when you visit your doctor – the list is endless.

“Data helps you react fast to consumer needs and helps companies address them faster,” says Partha De Sarkar, CEO of Hinduja Global Solutions. He says statistical modelling, thanks to modern data libraries and computing power, has combined with AI and Machine Learning algorithms to throw up insights about a customer like never before.

AI is the beginning of a human-machine partnership but this partnership should start off with the coming together of many minds – sociologists, scientists, and engineers – who must deliberate on the effects of AI on communities and individuals.

“In the end, it is the treatment of the data where biases creep in,” says Varun Mayya, co-founder of Avalon Labs. He says every founder must be responsible for the AI platforms they up even before they get consumers and clients to use them.

Experiences make each person different, but that is not exactly how AI works. Its algorithms bucket humans in to different date types, disregarding cultures, and preferences.

The algorithm bias

The cognitive revolution, touted as the next best thing in AI, thus falls flat when engineers use data to typecast individuals in a data set. “It is important for those claiming to use AI for consumer services to work with psychologists and sociologists before claiming their systems are representative of all communities and races,” says Nischith Rastogi, co-founder of Locus, a logistics tech company.

One such example where machine learning models erred with biases is the underwriting of loans. The machine set higher rates for individuals who it thought came from certain communities, income brackets and geography, not taking into account individuals who had the ability to service a loan.

“Biases creep into AI fast. It is something that startups and corporates should be cognisant of,” says Nischith.

The question then is, why is data biased? It starts from the collection of this data and medium it is captured from – the smartphone.

Smartphones create billions of data points about our food habits, fitness regimens, conversations, shopping lists, and payments. Here are a few biases thrown in by AI –

Entertainment: When several members of a family together watch an online streaming service, recommendations are based on past selections. Now, these may be of a particular individual and not necessarily what would serve a common interest.

According to a blog by PWC there’s a need to understand the bias in data, the strengths of the algorithms used, and “generalisability” of unseen data.

The blog adds that while the governance structure used for standard statistical models can be used for machine learning, there are a number of additional elements of software development that must be considered. PWC continues to warn that the tests machine learning models “go through” need to be significantly more robust, and a machine learning governance quality assurance framework will make developers more aware of statistical and software engineering constructs that the model operates within.

According to IBM, AI systems are only as good as the data we put into them. Poor data can contain racial, gender, or ideological biases and many AI systems continue to be trained using bad data, making it an on-going problem. “But we believe that bias can be tamed and that the AI systems that will tackle bias will be the most successful,” says IBM in its blog.

Retail and fashion: If you shop for fashion or beauty products, there is not only peer pressure to contend with now, but that from AI recommendations as well. AI today tends to attack you with a plethora of choices. And with fashion comes its ugly cousin – body shaming!

Earlier, the written word, in the form of fashion magazines, carried bias with pictures, and now, the same biases are carried over when building AI recommendations. With younger individuals taking to smartphones, these ‘recommendations’ may lead to unreasonable expectations from oneself.

Food and life:  Everyone wants to live healthy and it is widely understood that cultural moorings play a big role in what one can and cannot eat. The world of food and nutrition apps, however, tends to standardise profiles into broad strokes, and fitting people in broad data buckets.

“We are training our data models to be as robust as possible when it comes to recommendations. The algorithms learn only if developers ask the right questions,” says Tushar Vashist, co-founder of Healthifyme.

No wonder then that governments are beginning to sit up and take notice, and action. The UK parliament has commissioned a study on AI and the ethics surrounding its applications. The House of Lords-appointed Committee to “consider the economic, ethical and social implications of advances in artificial intelligence”, was set up on June 29, 2017, and will seek answers to five key questions:

• How does AI affect people in their everyday lives, and how is this likely to change?
• What are the potential opportunities presented by Artificial Intelligence for the UK? How can these be realised?
• What are the possible risks and implications of Artificial Intelligence?  How can these be avoided?
• How should the public be engaged with in a responsible manner about AI?
• What are the ethical issues presented by the development and use of Artificial Intelligence?

There are also strong voices around the world on the need for regulatory bodies for Artificial Intelligence to study the ethics of AI.

Back home, what do Indian policymakers have to say about this? Nothing much, is the simple answer.

The Niti Aayog, which creates broad policy frameworks, is keen on creating opportunities for Indians to invest in AI, but is mum on the moral and ethical frameworks of the technology.

“We absolutely need auditability and explainability,” says K M Madhusudan, CTO of Mindtree. He adds there are two aspects to this – one is for serious enterprise-level AI adoption, for which technologists must ensure AI can explain why it made a particular decision. The second is to ensure it is not biased.

The list of biases can be endless. But, it’s time for startups using AI to wake up and smell reality. It is in their interest to do so because they will hopefully soon be liable for the instructions or recommendations made by the AI. To avoid this, one must venture into creating reams of data before providing choices to individuals. In the end we are just one big data set.

The post Will Artificial Intelligence override human intelligence and experiences? appeared first on Artificial Intelligence.

Voices for the need of an ethical Artificial Intelligence are getting louder. AI has a great many applications, and has been solving various problems for humans. But such a powerful technology raises equal concerns about its possible misuse. And there have already been multiple cases of AI being used for malicious purposes.

Take Deepfakes for example. It is an advanced AI software that creates an artificial version of a person’s face accurately matching that of another. The software has been abused to create fake pornographic content using facial data of Hollywood celebrities such as Gal Gadot, Arianna Grande and Taylor Swift, among others.

Just recently, MIT experimented with an AI that was fed with data from the “darkest corners of Reddit.” The result was the world’s first psychopath AI, Norman.

Another recent example is Google’s Duplex AI, a natural-sounding robo assistant that makes calls on your behalf. Even though Google is still developing this AI, people have raised concerns over privacy and possible misuse by marketers or political campaigns.

Perhaps this is why the likes of Elon Musk have long warned about an AI apocalypse. In the case of Musk, he has even gone on to say that AI can one day lead to World War 3.

So, where do we go from here? There are two options we can consider: Abandon AI completely and forget about any benefits it can deliver, or explore the solutions that can prevent a possible apocalypse. Although governments across the world are yet to formulate regulations for AI, technology firms are already working towards implementing an “ethical AI”.

Google is already setting the standard by pledging that it will not allow its AI technology to be used for weapons or combat. Google will not pursue AI applications in “technologies that cause or are likely to cause overall harm. Where there is a material risk of harm, we will proceed only where we believe that the benefits substantially outweigh the risks, and will incorporate appropriate safety constraints”, wrote CEO Sundar Pichai in a blog post.

“AI algorithms and datasets can reflect, reinforce, or reduce unfair biases. We recognise that distinguishing fair from unfair biases is not always simple, and differs across cultures and societies. We will seek to avoid unjust impacts on people, particularly those related to sensitive characteristics such as race, ethnicity, gender, nationality, income, sexual orientation, ability, and political or religious belief,” he added.

But Google is not alone in this crusade. Technology firms like Microsoft and Accenture are already making giant strides in the direction of ethical AI.

Brad Smith, the President of Microsoft who co-authored ‘The Future Computed: Artificial Intelligence and its role in society’ with Harry Shum, laid out six principles for the AI to consider – fairness; reliability and safety; privacy and security; inclusiveness; transparency; and accountability. According to Smith, just a consensus on ethical AI is not enough.

“Because if we have a consensus on ethics and that is the only thing we do, then what we are going to find is that only ethical people will design AI systems ethically. That will not be good enough … We need to take these principles and put them into law,” he said at an event during a visit to Singapore in April this year. “Only by creating a future where AI law is as important a decade from now as, say, privacy law is today, will we ensure that we live in a world where people can have confidence that computers are making decisions in ethical ways.”

Accenture is taking a slightly different approach by monitoring the data input at the very initial levels.

“Accenture’s ‘Teach and Test’ methodology ensures that AI systems produce the right decisions in two phases. The ‘Teach’ phase focuses on the choice of data, models and algorithms that are used to train machine learning. This phase experiments and statistically evaluates different models to select the best performing model to be deployed into production, while avoiding gender, ethnic and other biases, as well as ethical and compliance risks,” the company said in a note.

“The adoption of AI is accelerating as businesses see its transformational value to power new innovations and growth,” said Bhaskar Ghosh, group chief executive, Accenture Technology Services. “As organisations embrace AI, it is critical to find better ways to train and sustain these systems – securely and with quality – to avoid adverse effects on business performance, brand reputation, compliance and humans.”

The general outlook is that the onus is on humans to build an AI that meets ethical parameters.

“As AI algorithms are created and trained by humans, there is a very high possibility of human bias being built into these algorithms. AI systems are superior to humans in speed and capability which, when used in malicious ways, can cause damage that is much higher in magnitude. Just as in humans, mistakes are unavoidable for AI algorithms too. The issue is the lack of accountability that would otherwise be a deterrent to negative actions,” said Arjun Pratap, CEO & Founder, EdGE Networks, a HR Technology Solutions provider that uses Data Science and Artificial Intelligence.

Back in India, people have started taking AI seriously. Niti Aayog recently released a discussion paper on the technology. Titled “National Strategy for Artificial Intelligence”, the paper touched on the issue of ethics along with exploring the massive benefits of the AI.

“As AI-based solutions permeate the way we live and do business, questions on ethics, privacy and security will also emerge. Most discussions on ethical considerations of AI are a derivation of the FAT framework (Fairness, Accountability and Transparency). A consortium of Ethics Councils at each Centre of Research Excellence can be set up and it would be expected that all COREs adhere to standard practice while developing AI technology and products,” said the paper.

“Data is one of the primary drivers of AI solutions, and thus appropriate handling of data, ensuring privacy and security is of prime importance. Challenges include data usage without consent, risk of identification of individuals through data, data selection bias and the resulting discrimination of AI models, and asymmetry in data aggregation. The paper suggests establishing data protection frameworks and sectorial regulatory frameworks, and promotion of adoption of international standards,” it added.

Experts suggest that AI can have a positive impact, provided the government is able to implement the right regulations.

“In India, businesses have only started harnessing AI in the recent past. It is definitely an area India should focus on as AI can create a huge impact in a positive and ethical way – provided the Indian government implements the necessary regulations and standards to ensure its responsible usage. A team that audits and certifies AI systems on factors such as trust, safety, accountability would certainly help in keeping malicious use of AI under check. It would foster safe, explainable and unbiased use of AI,” said Arjun on the possible steps needed to have ethical AI in India.

The post Ethical Artificial Intelligence is the need of the hour appeared first on Artificial Intelligence.

Most of us approach healthcare with hesitation. There are often several deterrents to seeking medical care (let alone preventive care)—ease of access being a major one. According to the Indian Journal of Public Health(September 2017 edition), India had just 4.8 practicing doctors per 10,000 population. While this is expected to grow to 6.9 doctors per 10,000 people by 2030, the minimum doctor-to-patient ratio recommended by the World Health Organization (WHO) is 1:1000.

Can we have more doctors? That is easier said than done. Even if we can, with our education system focused on quantity over quality, churning out doctors in droves will not guarantee better quality medical services in our country. Besides, with most graduates preferring lucrative urban locations, many Indians still find themselves at great physical or economic distance from quality healthcare.

Essentially, what we need is to fill the gap between the needs of the plenty and the services of the few. In my view, artificial intelligence (AI) has the capability to enable solutions that form the critical middle layer of access—making healthcare accessible and affordable to a large population base at the same quality level irrespective of people’s social standing.

Let us see how AI is set to impact our health in the coming years.

Reactive versus proactive healthcare

The usual attitude towards healthcare is, “I’ll cross that bridge when I come to it”—a reactive rather than proactive approach to seeking medical intervention. However, that has started changing in recent times. Sensors in wearables such as smart watches and Fitbits are already equipped to deliver actionable feedback to apps in our phones and connect to our doctor’s clinic for diagnostic tests and medication prescriptions.

Today, for instance, Apple watches can detect a variety of heart diseases—including diabetes prediction with an 85% match in known cases. All this by using simple, non-invasive tech that is already available.

Interesting AI start-ups like SigTuple (digitized blood analysis), Niramai (thermal scans for breast cancer) and Ten3T (portable, easy-to-use electrocardiograms) are currently developing disruptive diagnostic solutions that will significantly bring down costs, while making physical distance a non-issue. They (and others in this area) do this via cloud-based linkages to hospitals and clinics, chatbots, smart apps and AI-enabled data analytics. This means that sensors, real-time tracking and analytics will enable us to take pre-emptive charge of our own health, helping us to live more aware, healthier, longer lives.

Eliminating human biases

Barring simple ailments, most health consultations and treatments today come with some human bias. Sometimes, there are doubts if doctors’ or pharma companies’ vested interests are pushing certain treatments and medicines. That is why we gravitate towards known doctors. And for serious illnesses and critical care, second opinions are always recommended. In this context, AI-enabled medical care can save time, effort and costs through easy access to unbiased, consistent, good-quality diagnosis and treatment.

The experience of doctors also determines the options they explore. AI makes it possible to access the learnings and data from hundreds of thousands of cases. Oncology, for example, is an area where doctors are continually combining and recombining drugs to treat cancers or overcome resistance to previously successful drugs. Already, AI algorithms are helping doctors analyse a much wider scope of data and predicting—with greater granularity—new drug combinations that are personalized for a patient’s specific need.

Democratizing healthcare for 1.3 billion

Since technology can provide the middle layer bridge, AI-led systems have the potential to take healthcare to people irrespective of their location and affordability. People who live in rural or far-flung locations no longer have to be deprived of the up-to-date care offered at the nerve-centres of medical research. The tech increases the accessibility manifold. We thus have an opportunity to make this world truly equal.

Healthcare today—not just in India but pretty much the world over—is riddled with inefficiencies, and it’s easy to complain about them. However, we are poised for change, and technology has the potential to bridge the chasm between quality care and the many that need it but currently can’t access or afford that care. The beginnings have been made, and the work is well underway—significantly so in India—towards creating a level-field in healthcare.

The post Artificial Intelligence can make healthcare more accessible, affordable appeared first on Artificial Intelligence.

2017 was a formative year for artificial intelligence – not just in terms of the advancement of the technology itself, but also for the evolution of our understanding of AI’s impact on our society. Here are my predictions on the two topics that will become the primary focuses of AI in 2018.

Algorithms will change the way we work

When we talk about AI, often we focus on the belief that it is going to take away certain types of jobs (like cashiers or janitorial services). However, we often ignore the fact that this technology will also slowly permeate into most of our lives during work.

I refer to this as the distinction between AI and IA — artificial intelligence vs. intelligence augmentation. Despite the strides we’ve made in AI development and will continue to make in 2018, we are still years away from AI fully replacing human jobs. However, we are much closer to seeing the impact of AI permeate into almost every job and augment human intelligence.

Think about how you spend your days at work. There are likely parts of your day that require repetitive tasks or pattern recognition that could be delegated to someone without as much context and creativity. Take a doctor for example — AI will soon be able to detect and diagnose common diseases more quickly and accurately than humans. Imagine a world where computers read in a patient’s medical history, test results, and scans, and they are able to present doctors with various possible diagnoses and the likelihood that each of them is correct. Now, doctors can spend more of their time interpreting the possible diagnoses, communicating with patients, and developing unique, sustainable treatment plans that are most effective for each specific patient.

A world where doctors are replaced by AI — if that ever happens — is far off. However, a world where doctors are more efficient and effective with the help of AI is coming very soon. In cases like these, AI isn’t replacing anyone’s job; it’s simply allowing people to do their job better.

AI will be held against more scrutiny

AI is incredibly powerful and its adoption will only accelerate as it begins to augment our work and allow us to focus on the parts of our job that are most important. But part of why AI is so attractive is also why it’s so dangerous.

Algorithms are desirable for their ability to make well-informed decisions quickly and accurately. But that same power allows algorithms that are making incorrect decisions (and perpetuating biases) to do so with greater velocity and widespread impact than humans have ever had. As I mentioned in my earlier post about the Dark Side of Artificial Intelligence, there is potential for homogenous data sets to produce biased algorithms that are making important decisions about people’s lives — and we have already started to see the impact of these today (see Microsoft’s Twitter bot “Tay” and Google’s image results for Gorilla).

This is why, as AI becomes more prolific in 2018, it will also become more closely audited and scrutinized. Unfortunately, we will probably endure a few more experiences detailing what happens when AI is left unchecked. However, 2018 will be the year that companies purchasing AI products don’t just ask about the predictive power of an algorithm — they will stipulate that algorithms are tested in advance, interrogating and minimizing their potential unexpected impact.

The post Predictions For Artificial Intelligence In 2018 appeared first on Artificial Intelligence.

The field of artificial intelligence goes back a long way, but many consider it was officially born when a group of scientists at Dartmouth College got together for a summer, back in 1956. Computers had, over the last few decades, come on in incredible leaps and bounds; they could now perform calculations far faster than humans. Optimism, given the incredible progress that had been made, was rational. Genius computer scientist Alan Turing had already mooted the idea of thinking machines just a few years before. The scientists had a fairly simple idea: intelligence is, after all, just a mathematical process. The human brain was a type of machine. Pick apart that process, and you can make a machine simulate it.

The problem didn’t seem too hard: the Dartmouth scientists wrote, “We think that a significant advance can be made in one or more of these problems if a carefully selected group of scientists work on it together for a summer.” This research proposal, by the way, contains one of the earliest uses of the term artificial intelligence. They had a number of ideas—maybe simulating the human brain’s pattern of neurons could work and teaching machines the abstract rules of human language would be important.

The scientists were optimistic, and their efforts were rewarded. Before too long, they had computer programs that seemed to understand human language and could solve algebra problems. People were confidently predicting there would be a human-level intelligent machine built within, oh, let’s say, the next twenty years.

It’s fitting that the industry of predicting when we’d have human-level intelligent AI was born at around the same time as the AI industry itself. In fact, it goes all the way back to Turing’s first paper on “thinking machines,” where he predicted that the Turing Test—machines that could convince humans they were human—would be passed in 50 years, by 2000. Nowadays, of course, people are still predicting it will happen within the next 20 years, perhaps most famously Ray Kurzweil. There are so many different surveys of experts and analyses that you almost wonder if AI researchers aren’t tempted to come up with an auto reply: “I’ve already predicted what your question will be, and no, I can’t really predict that.”

The issue with trying to predict the exact date of human-level AI is that we don’t know how far is left to go. This is unlike Moore’s Law. Moore’s Law, the doubling of processing power roughly every couple of years, makes a very concrete prediction about a very specific phenomenon. We understand roughly how to get there—improved engineering of silicon wafers—and we know we’re not at the fundamental limits of our current approach (at least, not until you’re trying to work on chips at the atomic scale). You cannot say the same about artificial intelligence.

Common Mistakes

Stuart Armstrong’s survey looked for trends in these predictions. Specifically, there were two major cognitive biases he was looking for. The first was the idea that AI experts predict true AI will arrive (and make them immortal) conveniently just before they’d be due to die. This is the “Rapture of the Nerds” criticism people have leveled at Kurzweil—his predictions are motivated by fear of death, desire for immortality, and are fundamentally irrational. The ability to create a superintelligence is taken as an article of faith. There are also criticisms by people working in the AI field who know first-hand the frustrations and limitations of today’s AI.

The second was the idea that people always pick a time span of 15 to 20 years. That’s enough to convince people they’re working on something that could prove revolutionary very soon (people are less impressed by efforts that will lead to tangible results centuries down the line), but not enough for you to be embarrassingly proved wrong. Of the two, Armstrong found more evidence for the second one—people were perfectly happy to predict AI after they died, although most didn’t, but there was a clear bias towards “15–20 years from now” in predictions throughout history.

Measuring Progress

Armstrong points out that, if you want to assess the validity of a specific prediction, there are plenty of parameters you can look at. For example, the idea that human-level intelligence will be developed by simulating the human brain does at least give you a clear pathway that allows you to assess progress. Every time we get a more detailed map of the brain, or successfully simulate another part of it, we can tell that we are progressing towards this eventual goal, which will presumably end in human-level AI. We may not be 20 years away on that path, but at least you can scientifically evaluate the progress.

Compare this to those that say AI, or else consciousness, will “emerge” if a network is sufficiently complex, given enough processing power. This might be how we imagine human intelligence and consciousness emerged during evolution—although evolution had billions of years, not just decades. The issue with this is that we have no empirical evidence: we have never seen consciousness manifest itself out of a complex network. Not only do we not know if this is possible, we cannot know how far away we are from reaching this, as we can’t even measure progress along the way.

There is an immense difficulty in understanding which tasks are hard, which has continued from the birth of AI to the present day. Just look at that original research proposal, where understanding human language, randomness and creativity, and self-improvement are all mentioned in the same breath. We have great natural language processing, but do our computers understand what they’re processing? We have AI that can randomly vary to be “creative,” but is it creative? Exponential self-improvement of the kind the singularity often relies on seems far away.

We also struggle to understand what’s meant by intelligence. For example, AI experts consistently underestimated the ability of AI to play Go. Many thought, in 2015, it would take until 2027. In the end, it took two years, not twelve. But does that mean AI is any closer to being able to write the Great American Novel, say? Does it mean it’s any closer to conceptually understanding the world around it? Does it mean that it’s any closer to human-level intelligence? That’s not necessarily clear.

Not Human, But Smarter Than Humans

But perhaps we’ve been looking at the wrong problem. For example, the Turing test has not yet been passed in the sense that AI cannot convince people it’s human in conversation; but of course the calculating ability, and perhaps soon the ability to perform other tasks like pattern recognition and driving cars, far exceed human levels. As “weak” AI algorithms make more decisions, and Internet of Things evangelists and tech optimists seek to find more ways to feed more data into more algorithms, the impact on society from this “artificial intelligence” can only grow.

It may be that we don’t yet have the mechanism for human-level intelligence, but it’s also true that we don’t know how far we can go with the current generation of algorithms. Those scary surveys that state automation will disrupt society and change it in fundamental ways don’t rely on nearly as many assumptions about some nebulous superintelligence.

Then there are those that point out we should be worried about AI for other reasons. Just because we can’t say for sure if human-level AI will arrive this century, or never, it doesn’t mean we shouldn’t prepare for the possibility that the optimistic predictors could be correct. We need to ensure that human values are programmed into these algorithms, so that they understand the value of human life and can act in “moral, responsible” ways.

Phil Torres, at the Project for Future Human Flourishing, expressed it well in an interview with me. He points out that if we suddenly decided, as a society, that we had to solve the problem of morality—determine what was right and wrong and feed it into a machine—in the next twenty years…would we even be able to do it?

So, we should take predictions with a grain of salt. Remember, it turned out the problems the AI pioneers foresaw were far more complicated than they anticipated. The same could be true today. At the same time, we cannot be unprepared. We should understand the risks and take our precautions. When those scientists met in Dartmouth in 1956, they had no idea of the vast, foggy terrain before them. Sixty years later, we still don’t know how much further there is to go, or how far we can go. But we’re going somewhere.

The post When Will We Finally Achieve True Artificial Intelligence? appeared first on Artificial Intelligence.

Driving sales takes a lot of hard work. It’s a very time-consuming process, where you have to research potential clients, pitch to prospects, and keep following up with them before you close a deal.

Anything that can automate this process will certainly help salespeople close more deals and grow their business faster. The recent developments in artificial intelligence (AI), coupled with the availability of big data and deep learning, have given rise to a few interesting ways that can make it easier to grow your sales.

1. Schedule Meetings

AI-based calendar management tools like x.ai and tevorai.com can help your sales team manage their calendars and appointments automatically. They analyze the calendars of other employees to find free time slots when you can set up a meeting, and they also suggest alternative times, in case of unexpected delays. This saves a lot of time every day and fosters efficient collaboration. You don’t have to waste human hours trying to get the right people together at the right time.

2. Make Notes

One of the most important, yet tedious, aspects of every sales meeting is to take notes. Whether you’re meeting your prospects in person or through an audio call or a videoconference, it’s essential to make detailed notes, so you can effectively follow up later. Tools like clarke.ai and wrapup.co provide a powerful AI-based listening and note-making system that enables you to confidently preside over your sales meetings without the fear of losing any important points. Some of these tools even analyze client behavior and engagement to help you personalize your future communications.

3. Predict Customer Behavior

Did you know that AI systems can even create a customer profile for your business, based on user habits and behavior? By taking into account factors such as gender, age, location, demographics, and sentiments, they can predict what people need or are looking to buy. Salespeople can use these insights to refine their sales pitches and convert more leads into customers.

4. Target Customers Efficiently

If your salespeople don’t target the right prospects, they won’t be able to close many deals. AI-based platforms like Albert.ai use big data to help businesses target a specific group of potential customers, instead of going after a large pool of people. This allows you to execute marketing campaigns more effectively. While a data analyst would take hours to sift through big data sets and find the right prospects, an AI-based tool can do it in a matter of seconds. By empowering data analysts with AI systems, you can quickly and accurately sort through this information and improve your sales performance.

5. Increase Engagement

Today, instead of showing news in chronological order, companies are using AI to show people what they’re interested in. By learning user behavior and interests, companies like Facebook are using AI algorithms to make news feeds smarter and more personalized. What does this mean for your salespeople? This provides businesses more opportunities to put themselves in front of their target audience by posting updates that are related to people’s interests, resulting in more sales opportunities.

6. Provide More Selling Time

AI-driven, voice-controlled digital sales assistants like People.ai make it easy to manage your sales data, while your salespeople can spend more time on the actual process of selling. From managing leads to updating your CRM to improving forecasting, a digital assistant can take care of it all. By eliminating the need to manually enter data, AI-based systems not only ensure better data accuracy, but also leave your sales representatives more time to convert leads.

7. Close More Deals

With AI, you can improve every step of your sales process. You can start by identifying the right prospects. Furthermore, you can use AI-powered CRM systems like Salesforce’s Einstein to get detailed insights, smart recommendations, and predictions. Such powerful information will empower your salespeople to target the right people, personalize their sales pitch, add more value to every sales meeting, and close more deals faster. As data management is completely handled by AI, it will reduce errors and boost efficiency. With the availability of error-free data and actionable insights, your sales team can develop more effective strategies that drive conversions.

AI systems have changed the way businesses operate, and can improve the sales team performance by intelligently automating many of your day-to-day tasks, while adding a personal touch to your prospect interactions. By combining human behavior and AI-based tools, you can certainly organize and even optimize your sales processes.

The post 7 Ways Artificial Intelligence Can Help Improve Sales appeared first on Artificial Intelligence.

2016 turned out to be a great year for Artificial Intelligence. It had all of our attention and no one seemed to be bothered. Many start-ups tried their luck with AI. Everyone started to talk about how Artificial Intelligence is going to change lives and how it’s the “big thing”. 2017 did not have any lesser AI hype as well. Samsung launched Bixby which claims to be fully voice control and is considered to be better than Siri and Cortana.

Regardless of being criticized widely, the limelight around AI continued to shine. Though, in-depth researches and assumptions have a totally different stance. Being underdeveloped from 1955, AI was supposed to be a lot more efficient and apparently, intelligent by now but to be honest, it has failed miserably. It looks like the modern AI is inclined towards leveraging the business sector by showing off its magical abilities of data mining. It is safe to say that AI is going places but not the ones it really wants to.

Let me tell you something, there is no NLP. Not at least in these “Intelligent Assistants”. Deep learning, NLP and NLU are the buzzwords we all have been hearing for more than two years but very few startups are actually using them. Even ML and AIML are still not in the mainstream labs and what we have in the mainstream are retrieval-based talking applications that cannot learn and perform on their own will.

Machine Learning on the other hand, is the center of several kinds of research but the outcome is not really satisfying. Machine Learning is still light years away from achieving cognitive abilities of a human brain. The immense emotional drive and the dogmatic approach towards creation is still not present. Another thing that bothers me a lot is that Artificial Intelligence does not have any command over improvisation. If one door is closed, it does not bother knocking the other one and coming up with something totally out of the box i.e. a window.

Moving forward, we have also been hearing that “AI is going Mobile” and since more people are using mobile nowadays, it seems possible. But is it? Not right now at least. The main reason is that mobile AI is not in the right direction. Ever since Siri was launched, it was being said that Siri is the future of Artificial Intelligence and it’s going to become a major breakthrough in the tech scenario but what we found is that it’s more user focused and is actually NOT learning. The process that goes through is that every time we use Siri and every time it fails to entertain us it records the failure and the next version comes out improved.

In this scenario, engineers and data scientists are helping it learn. It is not learning on its own. Same thing happens with Cortana but it is far better than Siri. Since Cortana has control over Microsoft’s big data junk helping it in improving its efficiency. On the other hand, Google Now failed miserably in making a difference. All these “AI-Powered” assistants do not actually have AI involved in then. They do not have Neural Networks incorporated in their systems and they certainly do not possess Deep Learning algorithms. However, Samsung is cooking something slightly different with Bixby. It is rumored that Bixby is not going to have versions unlike Siri and Cortana and it has AI algorithms involved.

As of now, Bixby is not performing up-to-the-mark but in the near future, we can expect something better from it. However, it falls under the same AI-Assistant category is a batchmate of Siri and Cortana. Integrating these NNs in these assistants and ultimately attaching them to the data centers within the AIML framework is a hard task but it is not impossible.

With 100 billion neurons, and one trillion connections between them, human brain the unarguably the most complex system ever made. It is being said that AI will able to develop its own consciousness by 2030 but the thing is that it seems really impossible knowing the limited capabilities of Neural Networks. However, much has been said about the performance of Deep Learning algorithms but we haven’t seen real progress in that sector as well. Attributes like creativity, wisdom and social skills are still not present in AI’s plate. Why? Because NLU is still incapable of understanding the context. It sure has made remarkable progress in processing content but meaningful context is still hard for NLU engines to understand. Even voice recognition is not capable of giving away the hidden context under the layers of every what, that, where and why. Moreover, before the internet and even before computers, human minds were capable of churning out brilliant solutions to complex problems but AI is solely based on data and computing power which prioritizes human brain because of its rich history of development and knowledge. So, we can expect to have true AI in future but that future is kind of… far.

The post AI and Mobile AI: Are We on the Right Path? appeared first on Artificial Intelligence.