Mediaboss Marketing

Artificial intelligence is getting cheaper, better at the tasks we assign it, and more widespreadbut concerns over bias, ethics, and regulatory oversight still remain. At a time when AI is becoming accessible to everyone, the Stanford Institute for Human-Centered Artificial Intelligence put together a sweeping 2022 report analyzing the ins and outs of the growing field. Here are some of the highlights.

The number of publications alone on the topic tell a story: They doubled in the last decade, from 162,444 in 2010 to 334,497 in 2021. The most popular AI categories that researchers and others published on were pattern recognition, machine learning, and algorithms.

Whats more, the number of patent filings related to AI innovations in 2021 is 30 times greater than the filings in 2015. In 2021, the majority of filed patents were from China, but the majority of patents actually granted were from the US.

The number of users participating in open-source AI software libraries on GitHub also rose from 2015 to 2021. These libraries house collections of computer codes that are used for applications and products. One called TensorFlow remains the most popular, followed by OpenCV, Keras and PyTorch (which Meta AI uses).

Specifically, out of the various tasks that AI can perform, last year, the research community was focused on applying AI to computer vision, a subfield that teaches machines to understand images and videos in order to get good at classifying images, recognizing objects, mapping the position and movement of human body joints, and detecting faces (with and without masks).

[Related: MIT scientists taught robots how to sabotage each other]

For image classification, the most popular database used to train AI models is called ImageNet. Some researchers pre-train their models on additional datasets before exposing them to ImageNet. But models still make mistakes, on average mis-identifying 1 out of 10 images. The model that performs the best is from the Google Brain Team. In addition to identifying images and faces, AI can also generate fake images that are nearly indistinguishable from real ones, and to combat this, researchers have been working on deepfake detection algorithms that are based on datasets like FaceForensics++.

[Related: This new AI tool from Google could change the way we search online]

Natural language processing, a subfield that has been actively explored since the 1950s, is slowly making progress in English language understanding, summarizing, inferring reasonable outcomes, identifying emotional context, speech recognition and transcription, and translation. For basic reading comprehension, AI can perform better than humans, but when language tasks get more complicated, like when interpreting context clues is necessary, humans still have an edge. On the other hand, AI ethicists are worried that bias could affect large language models that draw from a mixed bag of training data.

Tech companies like Amazon, Netflix, Spotify, and YouTube have been improving the AI used in recommendation systems. The same is true for AIs role in reinforcement learning, which has enabled it to react and perform well in virtual games such as chess and Go. Reinforcement learning can also be used to teach autonomous vehicles tasks like changing lanes, or help data models predict future events.

As AI appears to have become better at doing what we want it to do, the cost to train it has come down as well, dropping by over 60 percent since 2018. Meanwhile, a system that wouldve taken 6 minutes to train in 2018 would now only take a little over 13 seconds. Accounting for hardware costs, in 2021, an image classification system would take less than $5 to train, whereas that cost wouldve been over $1,000 in 2017.

More AI applications across industries means more demand for AI education and jobs. Across the US in 2021, California, Texas, New York, and Virginia had the highest demand for AI-related occupations. In the last decade, the most popular specialties among PhD computer science students were artificial intelligence and machine learning.

Private investment in AI is at an all-time high, totalling $93.5 billion in 2021 (double the amount from 2020). AI companies that were skilled in data management, processing, and cloud, according to the report, got the most funding in 2021, followed by companies dedicated to medical and healthcare and financial technology (fintech for short).

In fiscal year 2021, US government agencies spent $1.53 billion on AI research and development for non-defense purposes, which was 2.7 times the amount spent in fiscal year 2018. For defense purposes, the Department of Defense allocated $9.26 billion across 500 AI research and development programs in 2021, which was about 6 percent more than what it spent in the year before. The top two uses of AI were for prototyping technologies and in programs countering weapons of mass destruction.

Last, the report looked at global, federal, and state regulations related to AI (looking for keywords like artificial intelligence, machine learning, autonomous vehicle or algorithmic bias). The report examined 25 countries around the world, and found that they have collectively passed 55 AI-related bills to law from 2016 to 2021. Last year, Spain, the UK and the US each had three AI-related bills that became law.

Continued here:
Artificial intelligence is everywhere now. This report shows how we got here. - Popular Science

MOOREA, French Polynesia. Titouan Bernicot was 16 when he had an epiphany: Everything he loved about living on an island in the South Pacific was tied to coral reefs.

Titouan Bernicot [Photo: Ryan Borne/Coral Gardeners]Moorea, where he lives, is made up of a coral reef and lagoon system, and even though the corals define the island, theyre increasingly at risk

Bernicot, 23, started Coral Gardeners in April 2017 from his bedroom. Now, he and his team are setting their sights on an audacious goal: to replant 1 million corals across the world by 2025. If they have any hope of hitting that number, theyll have to dramatically increase their planting, an effort that is being aided by a new artificial intelligence system.

Our goals for the next couple of years are to become the largest and most advanced coral restoration program on the planet, he said in an interview outside his office on the island.

[Photo: Ryan Borne/Coral Gardeners]The Coral Gardeners team restores damaged reef areas that have suffered from rising temperatures or other natural and human disturbances. There are currently three active nursery sites in French Polynesia, which have close to 9,000 corals growing. The first restored reef is in the back garden of Bernicots home, in Maharepa.

According to Coral Gardeners data, 500 million people around the world directly rely on reefs for food, income, coastal protection, and more.

[Photo: Ryan Borne/Coral Gardeners]The group uses various restoration techniques, and they focus efforts on super corals to create more resilient reefs. Super corals are a resistant species that can survive in extreme water temperatures. The group identifies super corals in the lagoon, cuts or trims a small portion of them, and places them in the nurseries on ropes, tables, or trees. There, they are monitored for at least a year. Once theyre large enough, the grown fragments are trimmed and planted onto damaged areas of the reef, where they continue growing, making the reef more resilient as the corals go on to spawn.

[Photo: Ryan Borne/Coral Gardeners]In the site outside Bernicots home, people can see the activity through a livestream known as ReefOS. The artificial intelligence system is a combination of hardware and software that documents every aspect of the growing process. Theres also a 360-degree camera that takes pictures and labels the fish species as they repopulate the reef.

[Photo: Ryan Borne/Coral Gardeners]In 2021, Coral Gardeners partnered with Cornell University to create a bioacoustic AI model with the reef soundscape, which helps to better understand the sounds effect on the ecosystem. They also teamed up with the University of Hawaii to integrate advanced 3D mapping and photogrammetry techniqueswhich is the science of extracting information from photographsinto the monitoring process. By modeling the structure of the reef, the team can learn critical metrics such as the biomass and the coral growth rate. Using this technology to map the reefs structure before and after the restoration efforts, Coral Gardeners can better track their impact.

[Photo: Ryan Borne/Coral Gardeners]The new techniques will allow the team to measure and understand the impact theyre having on the corals, including the biodiversity of the ecosystem, the fish population, coral disease, and bleaching detection.

[Photo: Ryan Borne/Coral Gardeners]The ReefOS AI model, which is currently in Maharepa and Cooks Bay, will be deployed at every restored reef site around the world as the organization expands. We want to become the most advanced reef restoration around the world in terms of technology . . . and also be able to share everything that we learned and the data we collected in an open-source and collaborative way, says Karine Toumazeau, strategist for Coral Gardeners.

The organization is on track to replant about 15,000 corals in 2022, and while this is the equivalent to what theyve transplanted in the past four years combined, its still a far cry from their goal of 1 million. The reefs theyve replanted thus far have about a 96% survival rate, according to Toumazeau.

[Photo: Ryan Borne/Coral Gardeners]While they werent able to provide an average cost-per-coral, the overhead for the project can be expensive. The group has raised $1.8 million in the last four years, but they hope to open at least 20 international branches in the next few years, which would require significantly more funding. So far, the majority of the funding has come from a coral adoption program, donations, merchandising, and partnerships.

Social media has been a real boon for fundraising. With more than half a million followers on Instagram, Coral Gardeners is the most followed account in all of French Polynesia, Bernicot notessomething he takes pride in. This following has helped drive growth, as celebrities like Paris Hilton have publicly given gifts of coral. (She gave it in honor of Britney Spears birthday.)

[Photo: Ryan Borne/Coral Gardeners]We are not in an ecosystem where you can meet a lot of entrepreneurs, he said. There arent a lot of start-up mentors in the middle of the South Pacific Ocean. I dont know any other projects that have that reach globally.

Still, one of the ideas behind Coral Gardeners is to create opportunities for local islanders. A childhood friend of Bernicots and the friends brother were among the first on the payroll of the more than 20 staff members.

I dont want ocean conservation to be a part-time job, Bernicot said. I want people that love the ocean to be able to make a career saving their playgrounds.

Continue reading here:
How one group is using artificial intelligence to replant coral reefs - Fast Company

ML and AI are essential drivers of innovation and growth in all sectors, including education.

Machine Learning (ML) and Artificial Intelligence (AI) are essential drivers of innovation and growth in all sectors, including education.

While AI-powered technologies have been around for a while in EdTech, the sector has been sluggish in their acceptance. The pandemic, on the other hand, radically altered the scene, pushing educators to rely on tech for virtual instruction. Now, 86 percent of educators believe that technology should be an integral element of education. AI has the potential to improve both learning and teaching, assisting the education industry, simultaneously evolving to benefit both students and teachers.

Here is how AI can benefit both the students and the educators:

To be precise, a students sole purpose of going to an educational institute is to get a degree or credential demonstrating their expertise. AI can have a huge impact on students educational journeys by offering access to the relevant courses, enhancing contact with teachers, and allocating more time to work on other aspects of life. Here are a few examples:

Personalization is one of the most prominent educational trends. Students now have a customized way of learning programs that focus on their own distinct experiences and interests; thanks to AI applications. AI can adjust to each students level of expertise, learning speed, and desired goals to ensure they get the most from their learning. Furthermore, AI-powered systems can examine students previous educational histories, detect shortcomings, and recommend courses better suited for improvement, allowing for a highly personalized learning opportunity.

While it is not unusual for kids to require additional assistance outside of the class, many educators would not have the time to assist children after school. While no chatbot can really replace a teacher, AI programs can assist students in honing their skills outside the classroom by helping with improving on weak areas. They offer one-on-one experiential learning without the teacher being available to answer questions at all hours of the day. In addition, an AI-powered bot can respond to queries in 2.7 seconds.

Nothing is more aggravating than posing a question and having it answered 2 days later. On a regular basis, teachers and instructors are assaulted with repetitious queries. With the support of automation and cognitive intelligence, AI can assist students to get solutions to their most frequently asked questions in seconds. This not only saves teachers a lot of time but also students time looking for answers or awaiting a response to their inquiries.

AI-powered solutions make learning available to all students, at any time and from any location. Each learner has his own pace, and having 24/7 access allows kids to experiment with what works best for them without having to wait for an educator. Furthermore, students from all around the world can obtain high-quality learning without paying travel or living fees.

Most teachers and staff arent ashamed to say they battle with time management, which makes sense given the number of tasks on their daily to-do lists. By automating chores, assessing student performance, and eliminating the educational gap, AI can assist in freeing up teachers time. Heres how it works:

Like AI can customize learning education courses for students, it can also assist teachers in their work. AI can provide teachers with a clear image of subjects and courses which need revaluation, by studying students learning capacities and histories. This study enables teachers to design the most effective learning plan for every single student. By studying each students particular needs, teachers and lecturers can tailor their courses to meet the most prevalent knowledge gaps or issue areas before a learner falls far behind.

AI-powered chatbots with accessibility to a schools entire base of knowledge can answer a range of generic and repetitive inquiries students commonly have without having to contact a faculty member. This way, AI frees up time for teachers to concentrate on curriculum design, coursework research, and ways of increasing student engagement.

AIs potential to automate the most basic job includes tasks such as replacing administrative labour, grading papers, measuring learning patterns, responding to general questions, etc. A Telegraph poll found that teachers spend 31% of their time organizing courses, grading tests, and doing administrative duties. Teachers, on the other hand, can use support automation systems to automate manual tasks, giving themselves more time to concentrate on improving their teaching competency.

Share This ArticleDo the sharing thingy

See the article here:
What is the Role of Artificial Intelligence in the Education Sector? - Analytics Insight

After a long winter, the artificial intelligence field has seen a resurgence in the past 15 years as computer power increased and a lot of digital data became available. In the past few years alone, giant language models advanced so quickly to outpace benchmarks, computer vision capabilities took self-driving cars from the lab to the street, and generative models tested democracies during major elections.

But parallel to this technologys rapid rise is its potential for massive harm; technologists, activists, and academics alike began calling for better regulation and understanding of its impact.

This spring, Stanford Institute for Human-Centered AI (HAI) will address three of the most critical areas of artificial intelligence during a one-day conference free and open to all:

Stanford HAI Associate Director and linguistics and computer science professor Christopher Manning, who will cohost the event with HAI Denning Co-director and computer science professor Fei-Fei Li, explains what this conference will cover and who should attend.

This conference will look at key advances in AI. Why are we focusing on foundation models, accountable AI, and embodied AI? What makes these the areas where you expect major growth?

An enormous amount of work is going on in AI in many directions. For a one-day event, we wanted to focus in on a small number of areas that we felt were key to where the most important and exciting research might appear this decade. We ended up focusing on three areas. First, there has been enormous excitement and investment around the development of large pre-trained language models and their generalization to including multiple data modalities that we have named foundation models. Second, there has been an exciting resurgence of work linking AI and robotics, often enabled by the use of simulated worlds, which allow the exploration of embodied AI and grounding. Finally, the increasing concerns about understanding AI decisions and maintaining data privacy in part demand societal and regulatory solutions, but they are also an opportunity for technical AI advances as to how you can produce interpretable AI systems or systems that still work effectively on data that is obscured for privacy reasons.

Who are you excited to hear from?

Ilya Sutskever has been one of the central people at the heart of the resurgence of deep learning-based AI, starting from his breakthrough work on the computer vision system AlexNet with Geoff Hinton in 2012. His impact has grown since he became the chief scientist of Open AI, which among other things has led in the development of foundation models. Im looking forward to hearing more about their latest models such as InstructGPT and what he sees lying ahead.

The recent successes in AI just would not have been possible without the amazing breakthroughs in parallel computing largely led by NVIDIA. Bill Dally is a leader in computer architecture, and, for the last decade, he has been the chief scientist at NVIDIA. He can give us powerful insights into the recent and future advances in parallel computing via GPUs but also insights into the broader range of vision, virtual reality, and other AI research going on at NVIDIA.

And Hima Lakkaraju is a trailblazing Harvard professor developing new strands of work in trustworthy and interpretable machine learning. When AI models are used in high-stakes settings, most times people would like accurate and reliable explanations of why the systems make certain decisions. One exciting direction in Himas work is in developing formal Bayesian models that can give reliable explanations.

Who should attend this conference?

Through a combination of short talks and panel discussions, were trying to achieve a balance between technical depth and accessibility. So on the one hand this conference should be of interest to anyone working in AI as a student, researcher, or developer, but beyond that we hope to be able to convey some of the excitement, results, and progress in these areas to anybody with an interest in AI, whether as a scientist, decision maker, or concerned citizen.

What do you hope your audience will take away from this experience?

I hope the audience will get a deeper understanding of how AI has been able to advance so quickly in the last 15 years, where it might go next, and what we should and shouldnt worry about. I hope people will take away the awesome powers of the huge new foundation models that are being built. But equally they will see why building a model from mountains of digital data is not sufficient, and we want to explore embodied AI models in a physical or simulated world that can learn more as babies learn. And finally, we will see something about how there is now a lot of exciting technical work underway to address the worries and downsides of AI that have been very prominently covered in the media in recent years.

Interested in attending the 2022 HAI Spring Conference? Learn more or register.

Link:
What's Next in Artificial Intelligence? Three Key Directions - Stanford HAI

Ilias Tagkopoulos,University of California, Davis

Nitin Nitin, University of California, Davis

The worlds food system is ripe for disruption in an unprecedented way. There are many challenges that we are facing both domestically and worldwide. According to the EPA, about a third of the food supply is lost, accounting for 15% of municipal solid waste and 2% of our energy use. Pesticide and herbicide use has increased more than 10% over the past 5 years while not everyone has access to high-quality food they can affordwhich is partially responsible for 40% of U.S. adults and 18% of our teenagers being obese, while one in eight families in America are hungry.

Many of these challenges are caused by inefficiencies in the food processing and distribution supply chain, which is a vital value-added step in our food system. The pieces of the puzzle are all there: ubiquitous sensors and devices that generate data with unprecedented volume, velocity and veracity; mature computational methods to make use of them; connected markets that can take advantage of these innovations at a global scale; and a need to transform antiquated, obsolete components of the current system, whether because of consumer demand for personalization and empowerment, or the need for global food safety and sustainability. Millions are spent every year in both the private and public sector to bring forth innovative solutions in capturing market preference, food safety, food security, provenance and traceability, all the while creating superior products that taste good, are good for your health and dont break the bank.

Recently, a strategic initiative by the National Science Foundation (NSF) and the U.S. Department of Agriculture (USDA) has led to the birth of the USDA/NSF AI Institute for Next Generation Food Systems (AIFS). AIFS is a $20 million collaboration between UC Davis, UC Berkeley, UC Agricultural and Natural Resources, University of Illinois and Cornell University, with the mission to develop and leverage transformative AI for the ethical production and distribution of safe, sustainable, nutritious food with fewer resources. With more than 50 faculty participating, AIFS aspires to use AI as the connective tissue that brings together the different segments of the food system, from molecular breeding and agricultural production to food processing and consumer nutrition.

Food safety risks are the leading causes of food recalls in the industry and a significant impact on health, social and economic aspects of our society. These risks range across diverse categories of food products, including dairy, meat, fresh produce and raw dry powders, while their main cause is contamination of food products with microbial pathogens in the harvesting and processing environments. The sources of these contaminations are diverse and often are not detected a priori with conventional testing. That is due to a lack of comprehensive sampling techniques and services that can provide accurate results in an inexpensive and timely manner.

This is where the trifecta of next-generation sequencing, artificial intelligence and cyber-physical systems can have a multiplier effect in keeping our food supply safe. Traditional 16s gene sequencing and, more recently, metagenomics sequencing, together with rapid identification of the microbial consortia in a sample, can quickly detect the presence of dangerous pathogens, such as Listeria monocytogenes.

AI-driven algorithms can be trained to assess the outbreak risk level by calculating the relative abundances of the various microbes in factories and food processing facilities. This can eventually lead to an always-on alert and recommendation system that can predict potential contamination and recommend corrective actions to reduce the risk of outbreaks and the cost of product recalls.

Furthermore, digital twins can serve as a clone digital replica of the factories, distribution centers and other areas where contamination is possible, providing a platform to evaluate hypotheses, optimize solutions and better understand system dynamics while constantly integrating feedback from the physical system.

Eventually, AI-driven systems that have the flexibility to integrate the existing food microbial ecology, chemometric and physical data sets for a comprehensive assessment of food safety risks can revolutionize food safety and create a safer food system, from farm to fork.

The AI Institute for Food Systems Value Chain

AI enrichment through the food supply chain.

Many food processing operations such as sterilization of food products, drying and baking require significant energy and water resources. In addition, sanitation operations required for the hygiene of the processing equipment use a significant amount of energy, water and chemical resources. With emerging climate challenges, there is a marked need to develop solutions to address these challenges.

Many of these efforts have focused on conventional engineering approaches such as waste heat recovery and reuse of spent water resources. Digital twins can again help by becoming an inexpensive testbed, as they can provide a digital replica of a processing operation and enable real-time analysis of water, energy and chemical usage in a facility. With cloud computing and scientific computation techniques, practitioners can run millions of simulations in minutes to identify the parameters that lead to the best possible resource use. Adaptive techniques such as active learning can be used to incorporate feedback from the physical system and improve the systems performance in maximizing efficiency.

In addition to opportunities for process optimization, these combinations of AI and digital twin technologies can aid in process validation and verification. Process validation and verification are required by regulatory organizations to ensure the safety of food products. Validation and verification processes in the food industry often require inoculation of food with a surrogate microbe to target pathogens and its testing following processing. These are resource- and time-intensive processes. AI-enabled digital twin technologies and data analytics can provide real-time validation and verification of processing operations.

In addition, significant early-stage efforts have been made to AI solutions for quality evaluation of input and output streams from food processing operations. The development of AI-guided sorting of fresh produce, such as blueberries, has shown significant improvement in efficiency and reducing labor-intensive practices in the industry. There are opportunities to advance vision and sensor-guided sorting of input and output streams of diversity of food products to improve the quality of the products and reduce food waste. Despite these early successes, there are many other areas in the quality control of food processing operations that are still managed based on empirical human decision-making processes: e.g. consistency of pastes and juices derived from agricultural commodities, such as apples and tomatoes, which are largely managed based on human judgment.

AI can revolutionize food safety and create a safer food system.

Producing and making available food that is good for our health, wallet and taste palate is the Holy Grail of any food and nutrition company. AI can play a significant role here, both unlocking the mysteries of chemical composition of food and creating new functional products. A significant blocker for any AI chef is the lack of the molecular atlas of food, knowing at high resolution what is in each ingredient beyond the protein, fat and carbohydrate content that we have been used to for generations. Not all proteins or carbs have been born equal when it comes to what they do to our body and mind. Even when we know the compoundsmay they be small molecules, peptides, glycans or anything elsethat confer benefits and help us transition to a healthier state, each of us has different genetics and gut microbiota, which in turn lead to strikingly different responses.

The complexity of the food-host interaction is both fascinating and daunting, and this is exactly the space where visionary initiatives can have a transformative impact in our way of life. AI-driven product formulations that are tailored for the individual needs of target groups, and that are specific enough based on co-morbidities, age or biomarkers, can be a paradigm shift that will epitomize the Hippocratic Let food be thy medicine. Optimal food processing is an equally important and complementary task, as the texture, nutritional value and function of the final product depends on it, with many of those beneficial compounds being lost in the process.

In the next 5 years, we will witness a paradigm shift in how we perceive, produce and consume food. There are already significant efforts to adopt and adapt the latest in sensor technology and AI in various aspects of the food system, while a number of initiatives with significant funding have been focused on mapping the molecular composition of various food ingredients. The AI community has made major advances in creating explainable and interpretable AI solutions, with a focus on fairness, trustworthiness and the ability to predict even in areas where there is a scarcity of data. Taken together, we expect an adaptive radiation of solutions and a rich ecosystem of partnerships fueled by unprecedented innovation and a strong desire to bring forth the next generation of food systems for a better tomorrow.

Dr. Ilias Tagkopoulos is a professor of computer science and the Genome Center at the University of California, Davis, where he leads the Integrative Biology and Predictive Analytics laboratory. He also is the director of the USDA-NIFA/NSF AI Institute of Next-Generation Food Systems (AIFS), a seven-institute collaboration. His work addresses data integration, modeling, design and decision-making under uncertainly, with applications in clinical and nutrition data. He holds a MSc from Columbia and a Ph.D. from Princeton in electrical engineering.

Dr. Nitin Nitin is co-principal investigator and lead of the Food Processing and Distribution Cluster at the Artificial Intelligence Institute for Next Generation Food Systems (AIFS). He is a professor in the departments of food science and technology and biological and agricultural engineering at the University of California, Davis. His research focuses on improving food quality and safety by developing innovative solutions focused on food processing, encapsulation, novel antimicrobials, biosensors and imaging. He holds Ph.D.s in bioengineering and food engineering.

Read the original here:
Inside AI: Food Processing and Distribution in the Era of Artificial Intelligence - Inside Unmanned Systems