Mediaboss Marketing

The views expressed here are solely those of the author and do not necessarily represent the views of FreightWaves or its affiliates.

In this installment of the AI in Supply Chain series (#AIinSupplyChain), we explore how Pathmind, an early-stage startup based in San Francisco, is helping companies apply simulation and reinforcement learning to industrial operations.

I asked Chris Nicholson, CEO and founder of Pathmind, What is the problem that Pathmind solves for its customers? Who is the typical customer?

Nicholson said: The typical Pathmind customer is an industrial engineer working at a simulation consulting firm or on the simulation team of a large corporation with industrial operations to optimize. This ranges from manufacturing companies to the natural resources sector, such as mining and oil and gas. Our clients build simulations of physical systems for routing, job scheduling or price forecasting, and then search for strategies to get more efficient.

Pathminds software is suited for manufacturing resource management, energy usage management optimization and logistics optimization.

As with every other startup that I have highlighted as a case in this #AIinSupplyChain series, I asked, What is the secret sauce that makes Pathmind successful? What is unique about your approach? Deep learning seems to be all the rage these days. Does Pathmind use a form of deep learning? Reinforcement learning?

Nicholson responded: We automate tasks that our users find tedious or frustrating so that they can focus on whats interesting. For example, we set up and maintain a distributed computing cluster for training algorithms. We automatically select and tune the right reinforcement learning algorithms, so that our users can focus on building the right simulations and coaching their AI agents.

Echoing topics that we have discussed in earlier articles in this series, he continued: Pathmind uses some of the latest deep reinforcement learning algorithms from OpenAI and DeepMind to find new optimization strategies for our users. Deep reinforcement learning has achieved breakthroughs in gaming, and it is beginning to show the same performance for industrial operations and supply chain.

On its website, Pathmind describes saving a large metals processor 10% of its expenditures on power. It also describes the use of its software to increase ore preparation by 19% at an open-pit mining site.

Given how difficult it is to obtain good quality data for AI and machine learning systems for industrial settings, I asked how Pathmind handles that problem.

Simulations generate synthetic data, and lots of it, said Slin Lee, Pathminds head of engineering. The challenge is to build a simulation that reflects your underlying operations, but there are many tools to validate results.

Once you pass the simulation stage, you can integrate your reinforcement learning policy into an ERP. Most companies have a lot of the data they need in those systems. And yes, theres always data cleansing to do, he added.

As the customer success examples Pathmind provides on its website suggest, mining companies are increasingly looking to adopt and implement new software to increase efficiencies in their internal operations. This is happening because the industry as a whole runs on very old technology, and deposits of ore are becoming increasingly difficult to access as existing mines reach maturity. Moreover, the growing trend toward the decarbonization of supply chains, and the regulations that will eventually follow to make decarbonization a requirement, provide an incentive for mining companies to seize the initiative in figuring out how to achieve that goal by implementing new technology

The areas in which AI and machine learning are making the greatest inroads are mineral exploration using geological data to make the process of seeking new mineral deposits less prone to error and waste; predictive maintenance and safety using data to preemptively repair expensive machinery before breakdowns occur; cyberphysical systems creating digital models of the mining operation in order to quickly simulate various scenarios; and autonomous vehicles using autonomous trucks and other autonomous vehicles and machinery to move resources within the area in which mining operations are taking place.

According to Statista, The revenue of the top 40 global mining companies, which represent a vast majority of the whole industry, amounted to some 692 billion U.S. dollars in 2019. The net profit margin of the mining industry decreased from 25 percent in 2010 to nine percent in 2019.

The trend toward mining companies and other natural-resource-intensive industries adopting new technology is going to continue. So this is a topic we will continue to pay attention to in this column.

Conclusion

If you are a team working on innovations that you believe have the potential to significantly refashion global supply chains, wed love to tell your story at FreightWaves. I am easy to reach on LinkedIn and Twitter. Alternatively, you can reach out to any member of the editorial team at FreightWaves at media@freightwaves.com.

Dig deeper into the #AIinSupplyChain Series with FreightWaves:

Commentary: Optimal Dynamics the decision layer of logistics? (July 7)

Commentary: Combine optimization, machine learning and simulation to move freight (July 17)

Commentary: SmartHop brings AI to owner-operators and brokers (July 22)

Commentary: Optimizing a truck fleet using artificial intelligence (July 28)

Commentary: FleetOps tries to solve data fragmentation issues in trucking (Aug. 5)

Commentary: Bulgarias Transmetrics uses augmented intelligence to help customers (Aug. 11)

Commentary: Applying AI to decision-making in shipping and commodities markets (Aug. 27)

Commentary: The enabling technologies for the factories of the future (Sept. 3)

Commentary: The enabling technologies for the networks of the future (Sept. 10)

Commentary: Understanding the data issues that slow adoption of industrial AI (Sept. 16)

Commentary: How AI and machine learning improve supply chain visibility, shipping insurance (Sept. 24)

Commentary: How AI, machine learning are streamlining workflows in freight forwarding, customs brokerage (Oct. 1)

Commentary: Can AI and machine learning improve the economy? (Oct. 8)

Commentary: Savitude and StyleSage leverage AI, machine learning in fashion retail (Oct. 15)

Commentary: How Japans ABEJA helps large companies operationalize AI, machine learning (Oct. 26)

Authors disclosure: I am not an investor in any early-stage startups mentioned in this article, either personally or through REFASHIOND Ventures. I have no other financial relationship with any entities mentioned in this article.

See the rest here:
Commentary: Pathmind applies AI, machine learning to industrial operations - FreightWaves

This article was originally published here

J Oral Pathol Med. 2020 Nov 21. doi: 10.1111/jop.13135. Online ahead of print.

ABSTRACT

BACKGROUND/AIM: Machine learning analyses of cancer outcomes for oral cancer remain sparse compared to other types of cancer like breast or lung. The purpose of the present study was to compare the performance of machine learning algorithms in the prediction of global, recurrence-free five-year survival in oral cancer patients based on clinical and histopathological data.

METHODS: Data was gathered retrospectively from 416 patients with oral squamous cell carcinoma. The dataset was divided into training and test dataset (75:25 split). Training performance of five machine learning algorithms (Logistic regression, K-nearest neighbours, Nave Bayes, Decision tree and Random forest classifiers) for prediction was assessed by k-fold cross validation. Variables used in the machine learning models were age, sex, pain symptoms, grade of lesion, lymphovascular invasion, extracapsular extension, perineural invasion, bone invasion and type of treatment. Variable importance was assessed and model performance on the testing data was assessed using receiver operating characteristic curves, accuracy, sensitivity, specificity and F1 score.

RESULTS: The best performing model was the Decision tree classifier, followed by the Logistic Regression model (accuracy 76% and 60%, respectively). The Nave Bayes model did not display any predictive value with 0% specificity.

CONCLUSIONS: Machine learning presents a promising and accessible toolset for improving prediction of oral cancer outcomes. Our findings add to a growing body of evidence that Decision tree models are useful in models in predicting OSCC outcomes. We would advise that future similar studies explore a variety of machine learning models including Logistic regression to help evaluate model performance.

PMID:33220109 | DOI:10.1111/jop.13135

Link:
Comparison of machine learning algorithms for the prediction of five-year survival in oral squamous cell carcinoma - DocWire News

It wasnt too long ago that concepts such as communicating with your friends in real time through text or accessing your bank account information all from a mobile device seemed outside the realm of possibility. Today, thanks in large part to the cloud, these actions are so commonplace, we hardly even think about these incredible processes.

Now, as we enter the golden age of machine learning, we can expect a similar boom of benefits that previously seemed impossible.

Machine learning is already helping companies make better and faster decisions. In healthcare, the use of predictive models created with machine learning is accelerating research and discovery of new drugs and treatment regiments. In other industries, its helping remote villages of Southeast Africa gain access to financial services and matching individuals experiencing homelessness with housing.

In the short term, were encouraged by the applications of machine learning already benefiting our world. But it has the potential to have an even greater impact on our society. In the future, machine learning will be intertwined and under the hood of almost every application, business process and end-user experience.

However, before this technology becomes so ubiquitous that its almost boring, there are four key barriers to adoption we need to clear first:

The only way that machine learning will truly scale is if we as an industry make it easier for everyone regardless of skill level or resources to be able to incorporate this sophisticated technology into applications and business processes.

To achieve this, companies should take advantage of tools that have intelligence directly built into applications from which their entire organization can benefit. For example, Kabbage Inc., a data and technology company providing small business cash flow solutions, used artificial intelligence to adapt and help processquickly an unprecedented number of small business loans and unemployment claims caused by COVID-19 while preserving more than 945,000 jobs in America. By folding artificial intelligence into personalization, document processing, enterprise search, contact center intelligence, supply chain or fraud detection, all workers can benefit from machine learning in a frictionless way.

As processes go from manual to automatic, workers are free to innovate and invent, and companies are empowered to be proactive instead of reactive. And as this technology becomes more intuitive and accessible, it can be applied to nearly every problem imaginable from the toughest challenges in the information technology department to the biggest environmental issues in the world.

According to the World Economic Forum, the growth of AI could create 58 million net new jobs in the next few years. However, research suggests that there are currently only 300,000 AI engineers worldwide, and AI-related job postings are three times that of job searches with a widening divergence.

Given this significant gap, organizations need to recognize that they simply arent going to be able to hire all the data scientists they need as they continue to implement machine learning into their work. Moreover, this pace of innovation will open doors and ultimately create jobs we cant even begin to imagine today.

Thats why companies around the world such asMorningstar, Liberty MutualandDBS Bank are finding innovative ways to encourage their employees to gain new machine learning skills with a fun, interactive hands-on approach. Its critical that organizations should not only direct their efforts towards training the workforce they have with machine learning skills, but also invest in training programs that develop these important skills in the workforce of tomorrow.

With anything new, often people are of two minds: Either an emerging technology is a panacea and global savior, or it is a destructive force with cataclysmic tendencies. The reality is, more often than not, a nuance somewhere in the middle. These disparate perspectives can be reconciled with information, transparency and trust.

As a first step, leaders in the industry need to help companies and communities learn about machine learning, how it works, where it can be applied and ways to use it responsibly, and understand what it is not.

Second, in order to gain faith in machine learning products, they need to be built by diverse groups of people across gender, race, age, national origin, sexual orientation, disability, culture and education. We will all benefit from individuals who bring varying backgrounds, ideas and points of view to inventing new machine learning products.

Third, machine learning services should be rigorously tested, measuring accuracy against third party benchmarks. Benchmarks should be established by academia, as well as governments, and be applied to any machine learning-based service, creating a rubric for reliable results, as well as contextualizing results for use cases.

Finally, as a society, we need to agree on what parameters should be put in place governing how and when machine learning can be used. With any new technology, there has to be a balance in protecting civil rights while also allowing for continued innovation and practical application of the technology.

Any organization working with machine learning technology should be engaging customers, researchers, academics and others to determine the benefits of its machine learning technology along with the potential risks. And they should be in active conversation with policymakers, supporting legislation, and creating their own guidelines for the responsible use of machine learning technology. Transparency, open dialogue and constant evaluation must always be prioritized to ensure that machine learning is applied appropriately and is continuously enhanced.

Through machine learning weve already accomplished so much, and yet its still day one (and we havent even had a cup of coffee yet!). If were using machine learning to help endangered orangutans, just imagine how it could be used to help save and preserve our oceans and marine life. If were using this technology to create digital snapshots of the planets forests in real-time, imagine how it could be used to predict and prevent forest fires. If machine learning can be used to help connect small-holding farmers to the people and resources they need to achieve their economic potential, imagine how it could help end world hunger.

To achieve this reality, we as an industry have a lot of work ahead of us. Im incredibly optimistic that machine learning will help us solve some of the worlds toughest challenges and create amazing end-user experiences weve never even dreamed. Before we know it, machine learning will be as familiar as reaching for our phones.

Swami Sivasubramanianis vice president of Amazon AI, running AI and machine learning services for Amazon Web Services Inc. He wrote this article for SiliconANGLE.

Show your support for our mission with our one-click subscription to our YouTube channel (below). The more subscribers we have, the more YouTube will suggest relevant enterprise and emerging technology content to you. Thanks!

Support our mission: >>>>>> SUBSCRIBE NOW >>>>>> to our YouTube channel.

Wed also like to tell you about our mission and how you can help us fulfill it. SiliconANGLE Media Inc.s business model is based on the intrinsic value of the content, not advertising. Unlike many online publications, we dont have a paywall or run banner advertising, because we want to keep our journalism open, without influence or the need to chase traffic.The journalism, reporting and commentary onSiliconANGLE along with live, unscripted video from our Silicon Valley studio and globe-trotting video teams attheCUBE take a lot of hard work, time and money. Keeping the quality high requires the support of sponsors who are aligned with our vision of ad-free journalism content.

If you like the reporting, video interviews and other ad-free content here,please take a moment to check out a sample of the video content supported by our sponsors,tweet your support, and keep coming back toSiliconANGLE.

Visit link:
Before machine learning can become ubiquitous, here are four things we need to do now - SiliconANGLE News

Eventually the keyboard of a smartphone is able to complete a sentence that we started. Who ever, when starting a browser search, saw the software indicate exactly what they were looking for? This is the concept behind Machine Learning, a branch of artificial intelligence that aims to make systems learn to behave more intelligently based on a large amount of data.

While the idea of AI is to make machines in a certain way think like humans, Machine Learning automates processes, creating shortcuts and seeking to predict actions according to user behavior or by analyzing information from a multitude of sources.

As a system that behaves by analyzing data, Machine Learning uses the users information to create a line of learning according to the registered behavior. So you may ask yourself, when did I pass information about my behavior to the machine?. The answer is: every time you surf the internet, use online services or use a connected device.

Companies like Google, Microsoft and Amazon are responsible for much of the data traffic from various services, such as search engines, e-mail services and e-commerce. These companies have huge computational centers (Big Datas) and receive information about what people are looking for, talking about or even wanting to buy. This happens through algorithms that are able to analyze data from different sources, such as social networks, research histories and the like, and the machine can understand the users behavior and create different profiles according to location, age group and common interests.

Machine Learning is not simply automation, but understanding routines to establish a working pattern, for example: in a smart home, the owner leaves in the morning and always comes back around 6 pm; when you get home, the lights are turned on automatically and the coffee maker is turned on to make an afternoon coffee. What if the person arrives early so that they dont have to turn on the lights? And on a hot day, would it not be more interesting to have a drink or water instead of a coffee? This is exactly where Machine Learning can make a difference.

Analyzing the users behavior, a machine-learning system is able to only activate the house lights if necessary and can use the room temperature to consider whether it is more interesting to start the coffee maker or send a message to the owner recommending drinking more water in days with low humidity. All of this can be based on searches performed on the smartphone, trends in social networks and mainly cross-data.

In addition to browsers that indicate the best results according to navigation, Machine Learning is used in services such as streaming platforms that indicate content related to what you have recently watched, mobility apps that show the best path according to traffic flow and, of course, operating systems that are capable of creating assistants that behave more and more like real virtual secretaries.

Machine Learning is also very widespread in IT security systems and has more and more solutions that benefit from technological advances to implement machine learning in new areas, such as meteorology and medicine.

See the original post:
What is Machine Learning and what are its benefits - Somag News

Rare diseases are defined as having less than 200,000 patients in the U.S., and many rare diseases have less than 1,000 patients worldwide. Thus, collecting necessary data from all patients is critical to accurately understand the disease. Currently, patients must travel to select academic centers to be part of these studies. This becomes difficult for patients with complex medical needs, accommodation challenges, and for those who cannot losework time. These aspects reduce participant enrollment and retention. Costs to patient organizations for studies can exceed $10,000 per subject per year and participants typically do not have access to study results all limiting patients' access to and engagement in rare disease research.

Thanks to two years of planning and this grant from CZI, the FOXG1 Research Foundation is launching a groundbreaking study using technology and machine learning to supplement the current NHS model by digitally collecting and analyzing critical patient data in order to scale rare disease research without exponential cost. Most importantly, this model allows patients direct access to their consolidated, digitized data that also uniquely summarizes their experience, which can be used to get second opinions or share with multiple providers to facilitate and improve their care.

"In order to find cures for rare diseases, all aspects of drug development need to be democratized, especially the collecting, analyzing and sharing of patient data to better understand unknown diseases. This has to be easy for the patient, affordable for the advocacy group, and totally accessible for researchers," explains Nasha Fitter, CEO and cofounder of the FOXG1 Research Foundation.

The new digital Natural History platform is launching with four rare disease groups: FOXG1 syndrome (FRF), SLC13A5 deficiency (TESS Research Foundation), SYNGAP1-related disorder (SynGAP Research Fund) and Rett Syndrome (Rett Syndrome Research Trust). These advocacy groups are at the forefront of rare disease research and are dedicated to redefining the drug development process. For some of these groups, natural history studies already exist and this digital platform will augment the existing NH dataset and provide a valuable and unique service to families and researchers. Accumulating data on multiple rare disease groups also enables cross-referencing for potential therapies.

In a partnership with Ciitizen, a Palo-Alto medical records platform provider, the rare disease groups kicking off this new model will onboard patients (caregivers) to sign up on the platform that will digitally collect the patient's medical records on their behalf, and then the anonymized data will be extracted and available for clinicians, researchers and biopharma to aid in research and therapy development. Researchers will be able to access large amounts of data for these small diseases to help determine clinical endpoints for potential treatments.

Key benefits of this new Digital Natural History Study include:

"We're excited to support the FOXG1 Research Foundation's efforts as they spearhead innovative work to accelerate patient access and engagement in rare disease research," said Tania Simoncelli, Director of the Science in Society Program at CZI. "We expect this effort will produce learnings and applications relevant to the broader rare disease community."

The FOXG1 Research Foundation is proud to be a Chan Zuckerberg Initiative Rare As One Strategic Partner. For more information about the Rare As One Network please visit our webpage.

About the FOXG1 Research Foundation:Founded in 2017, the FRF is a 501(c) parent-led global organization dedicated to funding science along the path to a cure and therapies for children and adults who are afflicted with the severe, rare, neurodevelopmental genetic disorder called FOXG1 syndrome. FOXG1 syndrome is characterized by severe developmental, cognitive, and physical disabilities, and epilepsy. For more information, please visit http://www.foxg1research.org.

About the Chan Zuckerberg Initiative:Founded by Dr. Priscilla Chan and Mark Zuckerberg in 2015, CZI is a new kind of philanthropy that's leveraging technology to help solve some of the world's toughest challenges from eradicating disease, to improving education, to reforming the criminal justice system. Across three core Initiative focus areas of Science, Education, and Justice & Opportunity, we're pairing engineering with grant-making, impact investing, and policy and advocacy work to help build an inclusive, just and healthy future for everyone. For more information, please visit http://www.chanzuckerberg.com.

Media Contact:Nicole Johnson, Co-Founder, Communications Director, FOXG1 Research Foundation [emailprotected]

SOURCE FOXG1 Research Foundation; SynGAP Research Fund

https://syngapresearchfund.org/

See the original post here:
FOXG1 Research Foundation to Pioneer a Machine Learning Approach to Accelerate Rare Disease Research with Support From the Chan Zuckerberg Initiative...