Archive for the ‘Artificial Intelligence’ Category

Spotlight on AI: Latest Developments in the Field of Artificial Intelligence – Analytics Insight

Whats new in the world of artificial intelligence?

Artificial intelligence is changing the course of our lives with its constant developments. Before the pandemic and now in the new normal, AI remains to be a key trend in the tech industry. It is reaching wider audiences as years pass and scientists, engineers, and entrepreneurs who involve themselves with modern technologies are reaping the benefits of AI and its branches, IoT and machine learning.

Organizations that overlooked digital transformation and the power of artificial intelligence are picking the pace of AI adoption. When COVID-19 was creating chaos across industries, it became evident that disruptive technologies and the automation that comes with it are more than crucial.

While 2020 was a great year for artificial intelligence working with its true potential, here are the latest advancements in the field of AI that are promising exciting times for the future of this technology.

Researchers from the University of Gothenburg have found an artificial intelligence model to predict what kind of virus can possibly spread from animals to humans. Using artificial intelligence, the algorithm studies the role of carbohydrates to understand the infection path. In scientific terms, carbohydrates are called glycans and they play a significant role in the way our bodies function. Almost all viruses first affect the glycans in our bodies, so did the coronavirus. Led by Daniel Bojar, assistant professor at the University of Gothenburg, the new AI model can analyze glycans with improved accuracy. The model analyses the infection process by predicting new virus-to-glycan interactions to better understand zoonotic diseases.

The world is evolving with disruptive technologies and that includes hackers and cyber attackers. Cyberattacks have become more common amidst the remote working culture, where sensitive files and documents have become the prime targets. To deal with this pressing concern, V.S.Subrahmanian, a cybersecurity researcher at Dartmouth College, created an algorithm called Word Embedding-based Fake Online Repository Generation Engine (WE-FORGE) that generates fake patents that are under development. This makes it difficult for hackers to find what they are looking for. The system generates convincing fakes based on the keywords of a given document. For each keyword it identifies, it analyses a list of related topics and replaces the original keyword with one randomly chosen related word.

DataRobot announced its second major platform launch, DataRobot version 7.1 with new MLOps management agents, time series model enhancements, and automated AI reports. With an aim to provide lifecycle management for remote AI and machine learning models, DataRobots new launch will offer feature discovery push-down integration for Snowflake and time series Eureqa model improvements. Through this, Snowflake users can use automatic discovery and computation of individual independent variables in the Snowflake data cloud. Apart from these additions, DataRobot also provides a no-code app builder that has the ability to convert deployed models into AI apps with no coding.

Exscientias US$60M acquisition of Allcyte will boost AI-drug discovery. Allcyte is an Austrian company that is developing an artificial intelligence platform to study how cancer treatments work on different individuals. Post the acquisition, this technology will work with Exscientias native software that uses AI to identify potential drug targets, build the right drugs, and send them for trials. Exscientia will now be able to work with a precision medicine approach to design drug molecules, ensuring improved efficiency.

The creator of the SaaS delivery model for financial services billing solutions, Redi2 Technologies, has announced a collaboration with IBM Private Cloud Services to improve flexibility. The combination of these technologies will add strong value for Redi2 Revenue Manager clients. Top asset managers throughout the world can take advantage of the improvements like fast reaction options for clients who need quick responses to changes, move data from one country to another or expand their infrastructure requirements.

Share This ArticleDo the sharing thingy

Continued here:
Spotlight on AI: Latest Developments in the Field of Artificial Intelligence - Analytics Insight

Evolution, rewards, and artificial intelligence – TechTalks

This article is part of the philosophy of artificial intelligence, a series of posts that explore the ethical, moral, and social implications of AI today and in the future

Last week, I wrote an analysis of Reward Is Enough, a paper by scientists at DeepMind. As the title suggests, the researchers hypothesize that the right reward is all you need to create the abilities associated with intelligence, such as perception, motor functions, and language.

This is in contrast with AI systems that try to replicate specific functions of natural intelligence such as classifying images, navigating physical environments, or completing sentences.

The researchers go as far as suggesting that with well-defined reward, a complex environment, and the right reinforcement learning algorithm, we will be able to reach artificial general intelligence, the kind of problem-solving and cognitive abilities found in humans and, to a lesser degree, in animals.

The article and the paper triggered a heated debate on social media, with reactions going from full support of the idea to outright rejection. Of course, both sides make valid claims. But the truth lies somewhere in the middle. Natural evolution is proof that the reward hypothesis is scientifically valid. But implementing the pure reward approach to reach human-level intelligence has some very hefty requirements.

In this post, Ill try to disambiguate in simple terms where the line between theory and practice stands.

In their paper, the DeepMind scientists present the following hypothesis: Intelligence, and its associated abilities, can be understood as subserving the maximisation of reward by an agent acting in its environment.

Scientific evidence supports this claim.

Humans and animals owe their intelligence to a very simple law: natural selection. Im not an expert on the topic, but I suggest reading The Blind Watchmaker by biologist Richard Dawkins, which provides a very accessible account of how evolution has led to all forms of life and intelligence on out planet.

In a nutshell, nature gives preference to lifeforms that are better fit to survive in their environments. Those that can withstand challenges posed by the environment (weather, scarcity of food, etc.) and other lifeforms (predators, viruses, etc.) will survive, reproduce, and pass on their genes to the next generation. Those that dont get eliminated.

According to Dawkins, In nature, the usual selecting agent is direct, stark and simple. It is the grim reaper. Of course, the reasons for survival are anything but simple that is why natural selection can build up animals and plants of such formidable complexity. But there is something very crude and simple about death itself. And nonrandom death is all it takes to select phenotypes, and hence the genes that they contain, in nature.

But how do different lifeforms emerge? Every newly born organism inherits the genes of its parent(s). But unlike the digital world, copying in organic life is not an exact thing. Therefore, offspring often undergo mutations, small changes to their genes that can have a huge impact across generations. These mutations can have a simple effect, such as a small change in muscle texture or skin color. But they can also become the core for developing new organs (e.g., lungs, kidneys, eyes), or shedding old ones (e.g., tail, gills).

If these mutations help improve the chances of the organisms survival (e.g., better camouflage or faster speed), they will be preserved and passed on to future generations, where further mutations might reinforce them. For example, the first organism that developed the ability to parse light information had an enormous advantage over all the others that didnt, even though its ability to see was not comparable to that of animals and humans today. This advantage enabled it to better survive and reproduce. As its descendants reproduced, those whose mutations improved their sight outmatched and outlived their peers. Through thousands (or millions) of generations, these changes resulted in a complex organ such as the eye.

The simple mechanisms of mutation and natural selection has been enough to give rise to all the different lifeforms that we see on Earth, from bacteria to plants, fish, birds, amphibians, and mammals.

The same self-reinforcing mechanism has also created the brain and its associated wonders. In her book Conscience: The Origin of Moral Intuition, scientist Patricia Churchland explores how natural selection led to the development of the cortex, the main part of the brain that gives mammals the ability to learn from their environment. The evolution of the cortex has enabled mammals to develop social behavior and learn to live in herds, prides, troops, and tribes. In humans, the evolution of the cortex has given rise to complex cognitive faculties, the capacity to develop rich languages, and the ability to establish social norms.

Therefore, if you consider survival as the ultimate reward, the main hypothesis that DeepMinds scientists make is scientifically sound. However, when it comes to implementing this rule, things get very complicated.

In their paper, DeepMinds scientists make the claim that the reward hypothesis can be implemented with reinforcement learning algorithms, a branch of AI in which an agent gradually develops its behavior by interacting with its environment. A reinforcement learning agent starts by making random actions. Based on how those actions align with the goals it is trying to achieve, the agent receives rewards. Across many episodes, the agent learns to develop sequences of actions that maximize its reward in its environment.

According to the DeepMind scientists, A sufficiently powerful and general reinforcement learning agent may ultimately give rise to intelligence and its associated abilities. In other words, if an agent can continually adjust its behaviour so as to improve its cumulative reward, then any abilities that are repeatedly demanded by its environment must ultimately be produced in the agents behaviour.

In an online debate in December, computer scientist Richard Sutton, one of the papers co-authors, said, Reinforcement learning is the first computational theory of intelligence In reinforcement learning, the goal is to maximize an arbitrary reward signal.

DeepMind has a lot of experience to prove this claim. They have already developed reinforcement learning agents that can outmatch humans in Go, chess, Atari, StarCraft, and other games. They have also developed reinforcement learning models to make progress in some of the most complex problems of science.

The scientists further wrote in their paper, According to our hypothesis, general intelligence can instead be understood as, and implemented by, maximising a singular reward in a single, complex environment [emphasis mine].

This is where hypothesis separates from practice. The keyword here is complex. The environments that DeepMind (and its quasi-rival OpenAI) have so far explored with reinforcement learning are not nearly as complex as the physical world. And they still required the financial backing and vast computational resources of very wealthy tech companies. In some cases, they still had to dumb down the environments to speed up the training of their reinforcement learning models and cut down the costs. In others, they had to redesign the reward to make sure the RL agents did not get stuck the wrong local optimum.

(It is worth noting that the scientists do acknowledge in their paper that they cant offer theoretical guarantee on the sample efficiency of reinforcement learning agents.)

Now, imagine what it would take to use reinforcement learning to replicate evolution and reach human-level intelligence. First you would need a simulation of the world. But at what level would you simulate the world? My guess is that anything short of quantum scale would be inaccurate. And we dont have a fraction of the compute power needed to create quantum-scale simulations of the world.

Lets say we did have the compute power to create such a simulation. We could start at around 4 billion years ago, when the first lifeforms emerged. You would need to have an exact representation of the state of Earth at the time. We would need to know the initial state of the environment at the time. And we still dont have a definite theory on that.

An alternative would be to create a shortcut and start from, say, 8 million years ago, when our monkey ancestors still lived on earth. This would cut down the time of training, but we would have a much more complex initial state to start from. At that time, there were millions of different lifeforms on Earth, and they were closely interrelated. They evolved together. Taking any of them out of the equation could have a huge impact on the course of the simulation.

Therefore, you basically have two key problems: compute power and initial state. The further you go back in time, the more compute power youll need to run the simulation. On the other hand, the further you move forward, the more complex your initial state will be. And evolution has created all sorts of intelligent and non-intelligent lifeforms and making sure that we could reproduce the exact steps that led to human intelligence without any guidance and only through reward is a hard bet.

Many will say that you dont need an exact simulation of the world and you only need to approximate the problem space in which your reinforcement learning agent wants to operate in.

For example, in their paper, the scientists mention the example of a house-cleaning robot: In order for a kitchen robot to maximise cleanliness, it must presumably have abilities of perception (to differentiate clean and dirty utensils), knowledge (to understand utensils), motor control (to manipulate utensils), memory (to recall locations of utensils), language (to predict future mess from dialogue), and social intelligence (to encourage young children to make less mess). A behaviour that maximises cleanliness must therefore yield all these abilities in service of that singular goal.

This statement is true, but downplays the complexities of the environment. Kitchens were created by humans. For instance, the shape of drawer handles, doorknobs, floors, cupboards, walls, tables, and everything you see in a kitchen has been optimized for the sensorimotor functions of humans. Therefore, a robot that would want to work in such an environment would need to develop sensorimotor skills that are similar to those of humans. You can create shortcuts, such as avoiding the complexities of bipedal walking or hands with fingers and joints. But then, there would be incongruencies between the robot and the humans who will be using the kitchens. Many scenarios that would be easy to handle for a human (walking over an overturned chair) would become prohibitive for the robot.

Also, other skills, such as language, would require even more similar infrastructure between the robot and the humans who would share the environment. Intelligent agents must be able to develop abstract mental models of each other to cooperate or compete in a shared environment. Language omits many important details, such as sensory experience, goals, needs. We fill in the gaps with our intuitive and conscious knowledge of our interlocutors mental state. We might make wrong assumptions, but those are the exceptions, not the norm.

And finally, developing a notion of cleanliness as a reward is very complicated because it is very tightly linked to human knowledge, life, and goals. For example, removing every piece of food from the kitchen would certainly make it cleaner, but would the humans using the kitchen be happy about it?

A robot that has been optimized for cleanliness would have a hard time co-existing and cooperating with living beings that have been optimized for survival.

Here, you can take shortcuts again by creating hierarchical goals, equipping the robot and its reinforcement learning models with prior knowledge, and using human feedback to steer it in the right direction. This would help a lot in making it easier for the robot to understand and interact with humans and human-designed environments. But then you would be cheating on the reward-only approach. And the mere fact that your robot agent starts with predesigned limbs and image-capturing and sound-emitting devices is itself the integration of prior knowledge.

In theory, reward only is enough for any kind of intelligence. But in practice, theres a tradeoff between environment complexity, reward design, and agent design.

In the future, we might be able to achieve a level of computing power that will make it possible to reach general intelligence through pure reward and reinforcement learning. But for the time being, what works is hybrid approaches that involve learning and complex engineering of rewards and AI agent architectures.

Read the original post:
Evolution, rewards, and artificial intelligence - TechTalks

Address artificial intelligence threats, politicians told – Business in Vancouver

B.C. Information and Privacy Commissioner Michael McEvoy: There is much good that comes from advancing AI technologies but if the public is to haveconfidence in itsuse we must first ensure trust and transparency is built into its development" | Photo: Jeremy Hainsworth

Governments' increasing use of artificial intelligence (AI) technology and peoples inability to avoid official computer services present threats politicians must address with law, privacy watchdogs say.Regulatory intervention is necessary, the B.C. and Yukon ombudsman and information and privacy commissioners said in a report released June 17.

The regulatory challenge is deciding how to adapt or modernize existing regulatory instruments to account for the new and emerging challenges brought on by governments use of AI. The increasing automation of government decision-making undermines the applicability or utility of existing regulations or common law rules that would otherwise apply to and sufficiently address those decisions.

Just as fairness and privacy issues resulting from the use of AI in commercial facial recognition systems have been shown to have bias and infringe peoples privacy rights, government use of AI can have serious, long-lasting impacts on peoples lives and could create tension with the fairness and privacy obligations of democratic institutions, the report said.

And that, they said, undermines trust in government.

While we recognize that delivering public services through artificially intelligent machine-based systems can be appealing to public bodies for cost reasons, we are concerned if not done right, this perceived efficiency may come at the expense of important rights to fair treatment, said B.C. Ombudsperson Jay Chalke.

The late Prof. John McCarthy of the Massachusetts Institute of Technology and Stanford University coined the term AI. The report said McCarthys definition frames AI in terms of the development of machines that can perform tasks normally requiring human intelligence, such as visual perception, speech recognition, language translation and decision-making. Its a capacity to respond to challenges and opportunities based on inputs and goals.

Whats happening, the officials said, is that AI is replacing the judgement of human decision-makers in governments around the world. Such computer judgments could include predicting criminals recidivism rates, approving building permits, determining government program eligibility and deciding car insurance premiums.

There is much good that comes from advancing AI technologies but if the public is to haveconfidence in itsuse we must first ensure trust and transparency is built into its development, said B.C. Information and Privacy Commissioner Michael McEvoy.

The report said global spending on AI was US$12.4 billion in 2019 and is expected to reach US$232 billion by 2025. As part of Canadas national AI strategy, Ottawa has invested $355 million to develop synergies between retail, manufacturing, infrastructure and information and communications technology sectors to build intelligent supply chains through AI and robotics.

However, another challenge comes from people themselves and a desire for fast service that could put highly personal and private data at risk.The report said Peter Tyndall, former president of the International Ombudsman Institute and the ombudsman of the Republic of Ireland, has argued one of the biggest challenges facing independent oversight offices and core government alike is peoples expectation of speedy results and high levels of interactivity.

They expect to interact with public services as they do with Amazon or Facebook, to communicate as they do on WhatsApp, Tyndall said.

Other concerns highlighted in the report include the challenge of explaining to the public how decisions are made if algorithms are used, a lack of notice provided to people that these systems will be used in decision-making that impacts them and the absence of effective appeals from AI-generated decisions.

When our offices reviewed how AI is being used, we saw there is a real gap in uniform guidance, regulation and oversight that governs the use of AI, said Ombudsman and Information and Privacy Commissioner for Yukon Diane McLeod-McKay. We are hopeful that public bodies will carefully consider the guidance we are providing when they are using AI.

The report makes several recommendations aimed at public bodies delivering public services including:

the need for public bodies to commit to guiding principles for AI use;

the need for public bodies to notify an individual when an AI system is used to make a decision about them and describe how the AI system operates in a way that is understandable to the individual;

the need for government to promote capacity-building, co-operation and public engagement on AI;

a requirement for all public bodies to complete and submit an artificial intelligence fairness and privacy impact assessment for all existing and future AI programs for review by the relevant oversight body; and

the establishment of special rules or restrictions for the use of highly sensitive information by AI.

jhainsworth@glaciermedia.ca

@jhainswo

Go here to read the rest:
Address artificial intelligence threats, politicians told - Business in Vancouver

Artificial Intelligence ‘Creates’ Its Own PLAYABLE Version of ‘GTA 5’, And It Is FREAKY – Tech Times

Artificial intelligence is still considered a long way from its feared destructive capabilities, but one thing is for certain: big things start from small beginnings. And that small beginning might as well be this.

(Photo : Andrea Verdelli/Getty Images)SHANGHAI, CHINA - JUNE 18: Cutting edge applications of Artificial Intelligence are seen on display at the Artificial Intelligence Pavilion of Zhangjiang Future Park during a state organized media tour on June 18, 2021 in Shanghai, China.

Engadgetreports that a certain artificial intelligence managed to somehow "create" its own version of "GTA V," and also make it actually playable--at least a short stretch of it, that is. And all they did was make the AI "watch" a portion of gameplay.

YouTuber Harrison Kinsley, who goes by the name sentdex on the video platform, shareda videoof artificial intelligence achieving the technically impressive feat. Working with a collaborator named Daniel Kukiela, Kinsley used a program called GameGAN Neural Network to create the simulation.

GameGAN, according toits website, is an artificial intelligence-based program created to simulate game environments on the fly. Made by the NVIDIA Toronto AI Lab (at least according to its Google Results Page heading), GameGAN has actually done the same exact thing last year, only with a different game: Pac-Man.

Before it managed to create its own playable version of "GTA V," GameGAN made its own version of Pac-Man by watching another AI play through it. According toEngadget, artificial intelligence managed to essentially "develop" an entire video game in merely four days. The real Pac-Man, on the other hand, took over a year to make.

This also isn't really the first time thatAI has dabbled in game visuals. NVIDIA's DLSS and so-called AI upscaling that's turned the graphics of "GTA V" into photorealistic images already exist.

Read also:Artificial Intelligence to Hunt for Dark Energy Using this INSANELY POWERFUL Supercomputer

To achieve the feat, Kinsley needed some major help. No consumer-class computer would be able to deal with this kind of workload, so NVIDIA loaned him and his partner a $200,000 data center; essentially a small-scale supercomputer. This machine contained eight A100 GPUs, which are optimized for hardware-accelerated artificial intelligence, as well as two 64-core server CPUs from AMD.

(Photo : China/Barcroft Media via Getty Images)HANGZHOU, CHINA - JUNE 06 2021: Visitors stop by an AI server based on NVIDIA A100 chips at the 2021 Global Artificial Intelligence Technology Conference (GAITC2021) in Hangzhou in East China.

With the data center, Kinsley and Kukiela "trained" the GameGAN using actual gameplay of "GTA V." They let 12 simultaneous AI instances drive the same stretch of road, allowing the hardware to collect enough data to build its own world. The result was an amazing testament to the power of artificial intelligence, which is also a little bit freaky.

And the AI itself didn't just create an image that moved. It also rendered rudimentary 3D graphics in real-time. What this means is that as the car drove around, the shadow underneath it also moved relative to where the light source is. That's absolutely amazing since doing that convincingly inside a game engine would take years of hard coding. A machine managed to do that in barely a fraction of the time.

So does this mean that game developers will be replaced by artificial intelligence? No. Absolutely not. For now, AI can stick to doing things likebeating humans in DOTA 2, or being funnily dumb in broken, unfinished games.

Related: Super Mario Artificial Intelligence Learns How to Feel, Plays Own Game [Video]

This article is owned by Tech Times

Written by RJ Pierce

2021 TECHTIMES.com All rights reserved. Do not reproduce without permission.

Originally posted here:
Artificial Intelligence 'Creates' Its Own PLAYABLE Version of 'GTA 5', And It Is FREAKY - Tech Times

Artificial intelligence won’t replace the role of financial advisors, UBS CEO says – CNBC

LONDON One of the world's biggest wealth managers doesn't think artificial intelligence can replace the role of financial advisors.

Ralph Hamers, the CEO of UBS, said Wednesday that technologies like AI were better suited to handling day-to-day functions like opening an account or executing trades than advising clients.

"There is no added value for client advisors to be engaged in a process like that," Hamers told CNBC's Geoff Cutmore at the virtual CNBC Evolve Global Summit. "They're advisors. They should advise."

"Our financial advisors actually should be supported by the technology," Hamers said, adding that AI could be used to make sense of the research and other data that advisors don't have time for.

"That is what artificial intelligence can do, because even our client advisors can't read all the research that is there," he said. "Our client advisors can't comprehend all the product options that are out there."

Europe's banking industry has seen radical change over the last decade, with new entrants like Monzo, Revolut and N26 emerging to take on incumbents with slick, digital-only services.

Covid-19 has further accelerated digital transformation in the banking sector, with many lenders racing to move away from their aging IT systems to cloud-based technology. Some are partnering with tech companies like Microsoft, Amazon and Google, as well as fintech upstarts, to hasten the process.

Hamers said UBS is looking to adopt a "Netflix experience" where clients have access to a "dashboard" of different research and products to choose from.

"That's where things are going, and that's where UBS is making the next step, in terms of dealing with technology to deliver a much better service for our clients," he added.

More:
Artificial intelligence won't replace the role of financial advisors, UBS CEO says - CNBC