LONDON--(BUSINESS WIRE)--The deep learning market is expected to grow by USD 7.2 billion during 2020-2024, according to the latest market research report by Technavio. Request a free sample report

Deep learning is popularly used in machine learning, which involves the use of artificial neural networks with several degrees of layers. Moreover, massive volumes of digital data that is produced at an unprecedented rate across industries is widening the application area of deep learning. In the healthcare industry, deep learning applications are used in drug research and development. Also, deep learning helps in training machines to understand the complexities associated with languages such as syntax and semantics and generating appropriate responses. Other application areas of deep learning are fraud detection, visual recognition, logistics, insurance, and agriculture. Thus, the growing applications of deep learning are expected to drive market growth during the forecast period.

To learn more about the global trends impacting the future of market research, download a free sample: https://www.technavio.com/talk-to-us?report=IRTNTR41147

As per Technavio, the growing emphasis on cloud-based deep learning will have a positive impact on the market and contribute to its growth significantly over the forecast period. This research report also analyzes other significant trends and market drivers that will influence market growth over 2020-2024.

Deep Learning Market: Growing Emphasis On Cloud-Based Deep Learning

Cloud computing is considered an appropriate platform for deep learning as it provides support for scalability, visualization, and storage of vast amounts of structured and unstructured data. The use of cloud computing in deep learning allows the integration of large datasets for training algorithms. Moreover, cloud computing also allows deep learning models to scale efficiently and at a much lower cost. Thus, the popularity of cloud-based deep learning is increasing, which will have a positive impact on the growth of the market during the forecast period.

Increasing collaboration among vendors and the rising investments in deep learning will have a significant impact on the deep learning market growth during the forecast period, says a senior analyst at Technavio.

Register for a free trial today and gain instant access to 17,000+ market research reports

Technavio's SUBSCRIPTION platform

Deep Learning Market: Segmentation Analysis

This market research report segments the deep learning market by type (software, services, and hardware), and geographic segmentation (APAC, Europe, MEA, North America and South America).

North America region led the deep learning market in 2019, and the region is expected to register the highest incremental growth during the forecast period. This can be attributed due to factors such as the increasing use of deep learning in various industrial applications such as voice recognition and image recognition.

Technavios sample reports are free of charge and contain multiple sections of the report, such as the market size and forecast, drivers, challenges, trends, and more. Request a free sample report

Some of the key topics covered in the report include:

Type segmentation

Geographic segmentation

Market Drivers

Market Challenges

Market Trends

Vendor Landscape

About Technavio

Technavio is a leading global technology research and advisory company. Their research and analysis focus on emerging market trends and provides actionable insights to help businesses identify market opportunities and develop effective strategies to optimize their market positions.

With over 500 specialized analysts, Technavios report library consists of more than 17,000 reports and counting, covering 800 technologies, spanning across 50 countries. Their client base consists of enterprises of all sizes, including more than 100 Fortune 500 companies. This growing client base relies on Technavios comprehensive coverage, extensive research, and actionable market insights to identify opportunities in existing and potential markets and assess their competitive positions within changing market scenarios.

Originally posted here:

Global Deep Learning Market 2020-2024 | Growing Application of Deep Learning to Boost Market Growth | Technavio - Business Wire

Supervised Learning

Supervised machine learning builds a model that makes predictions based on evidence in the presence of uncertainty. A supervised learning algorithm takes a known set of input data and known responses to the data (output) and trains a model to generate reasonable predictions for the response to new data. Use supervised learning if you have known data for the output you are trying to predict.

Supervised learning uses classification and regression techniques to develop predictive models.

Classification techniques predict discrete responsesfor example, whether an email is genuine or spam, or whether a tumor is cancerous or benign. Classification models classify input data into categories. Typical applications include medical imaging, speech recognition, and credit scoring.

Use classification if your data can be tagged, categorized, or separated into specific groups or classes. For example, applications for hand-writing recognition use classification to recognize letters and numbers. In image processing and computer vision, unsupervised pattern recognition techniques are used for object detection and image segmentation.

Common algorithms for performing classification include support vector machine (SVM), boosted and bagged decision trees, k-nearest neighbor, Nave Bayes, discriminant analysis, logistic regression, and neural networks.

Regression techniques predict continuous responsesfor example, changes in temperature or fluctuations in power demand. Typical applications include electricity load forecasting and algorithmic trading.

Use regression techniques if you are working with a data range or if the nature of your response is a real number, such as temperature or the time until failure for a piece of equipment.

Common regression algorithms include linear model, nonlinear model, regularization, stepwise regression, boosted and bagged decision trees, neural networks, and adaptive neuro-fuzzy learning.

Continue reading here:

What Is Machine Learning? | How It Works, Techniques ...

This wide-ranging article identifies how regulators have slowly opened up to accept the use of machine learning models as a method of detecting AML activity, yet they remain concerned regarding the models lack of transparency. It reviews public comments made by key regulators regarding technology and the need to maintain balance between detection and inhibiting commerce and protecting privacy.

Here is one small part of the article that is well worth reading if you are interested in AML processing:

At a November, 2018, Fintech and the New Financial Landscape conference in Philadelphia Pennsylvania conference Dr. Lael Brainard presented her view about the potential for AI and machine learning. In short, while Dr Brainard is bullish on the transformative capabilities of AI and Machine Learning, she is cautious about explainability and the audit-ability of black box AI models. She states the need for guard-rails to contain AI risk, while observing safety and soundness and consumer financial protection.

In her address entitled What Are We Learning about Artificial Intelligence in Financial Services?, she told delegates she is optimistic about the potential for AI and machine learning in particular, but guarded on how new machine learning models can be audited.

Dr. Brainards well informed speech begins, Modern machine learning applies and refines, or trains, a series of algorithms on a large data set by optimizing iteratively as it learns in order to identify patterns and make predictions for new data. Machine learning essentially imposes much less structure on how data is interpreted compared to conventional approaches in which programmers impose ex ante rule sets to make decisions.

She accurately states the value of machine learning when applied to banking AML and loan processing; here are quotes from her remarks:

1.Firms view AI approaches as potentially having superior ability for pattern recognition, such as identifying relationships among variables that are not intuitive or not revealed by more traditional modeling.

2. Firms see potential cost efficiencies where AI approaches may be able to arrive at outcomes more cheaply with no reduction in performance.

3.AI approaches might have greater accuracy in processing because of their greater automation compared to approaches that have more human input and higher operator error.

4. Firms may see better predictive power with AI compared to more traditional approachesfor instance, in improving investment performance or expanding credit access.

5. AI approaches are better than conventional approaches at accommodating very large and less-structured data sets and processing those data more efficiently and effectively.

Dr. Brainard continues, The question is how should we approach regulation and supervision? It is incumbent on regulators to review the potential consequences of AI, including the possible risks, and take a balanced view about its use by supervised firms.Regulation and supervision need to be thoughtfully designed so that they ensure risks are appropriately mitigated but do not stand in the way of responsible innovations that might expand access and convenience for consumers and small businesses or bring greater efficiency, risk detection, and accuracy.

Overview byTim Sloane,VP, Payments Innovation at Mercator Advisory Group

Summary

Article Name

Regulators Begin to Accept Machine Learning to Improve AML but There Are Major Issues

Description

This wide ranging article identifies how regulators have slowly opened up to accept the use of machine learning models as a method of detecting AML activity yet remain concerned regarding the models lack of transparency.

Author

Tim Sloane

Publisher Name

PaymentsJournal

Publisher Logo

See the original post here:

Regulators Begin to Accept Machine Learning to Improve AML, But There Are Major Issues - PaymentsJournal

Payoneer uses Iguazio to move from detection to prevention of fraud with predictive machine learning models served in real-time.

Iguazio, the data science platform for real time machine learning applications, today announced that Payoneer, the digital payment platform empowering businesses around the world to grow globally, has selected Iguazios platform to provide its 4 million customers with a safer payment experience. By deploying Iguazio, Payoneer moved from a reactive fraud detection method to proactive prevention with real-time machine learning and predictive analytics.

Payoneer overcomes the challenge of detecting fraud within complex networks with sophisticated algorithms tracking multiple parameters, including account creation times and name changes. However, prior to using Iguazio, fraud was detected retroactively, enabling customers to only block users after damage had already been done. Payoneer is now able to take the same sophisticated machine learning models built offline and serve them in real-time against fresh data. This ensures immediate prevention of fraud and money laundering with predictive machine learning models identifying suspicious patterns continuously. The cooperation was facilitated by Belocal, a leading Data and IT solution integrator for mid and enterprise companies.

"Weve tackled one of our most elusive challenges with real-time predictive models, making fraud attacks almost impossible on Payoneer" noted Yaron Weiss, VP Corporate Security and Global IT Operations (CISO) at Payoneer. "With Iguazios Data Science Platform, we built a scalable and reliable system which adapts to new threats and enables us to prevent fraud with minimum false positives".

"Payoneer is leading innovation in the industry of digital payments and we are proud to be a part of it" said Asaf Somekh, CEO, Iguazio. "Were glad to see Payoneer accelerating its ability to develop new machine learning based services, increasing the impact of data science on the business."

"Payoneer and Iguazio are a great example of technology innovation applied in real-world use-cases and addressing real market gaps" said Hugo Georlette, CEO, Belocal. "We are eager to continue selling and implementing Iguazios Data Science Platform to make business impact across multiple industries."

Iguazios Data Science Platform enables Payoneer to bring its most intelligent data science strategies to life. Designed to provide a simple cloud experience deployed anywhere, it includes a low latency serverless framework, a real-time multi-model data engine and a modern Python eco-system running over Kubernetes.

Earlier today, Iguazio also announced having raised $24M from existing and new investors, including Samsung SDS and Kensington Capital Partners. The new funding will be used to drive future product innovation and support global expansion into new and existing markets.

About Iguazio

The Iguazio Data Science Platform enables enterprises to develop, deploy and manage AI applications at scale. With Iguazio, companies can run AI models in real time, deploy them anywhere; multi-cloud, on-prem or edge, and bring to life their most ambitious data-driven strategies. Enterprises spanning a wide range of verticals, including financial services, manufacturing, telecoms and gaming, use Iguazio to create business impact through a multitude of real-time use cases. Iguazio is backed by top financial and strategic investors including Samsung, Verizon, Bosch, CME Group, and Dell. The company is led by serial entrepreneurs and a diverse team of innovators in the USA, UK, Singapore and Israel. Find out more on http://www.iguazio.com

About Belocal

Since its inception in 2006, Belocal has experienced consistent and sustainable growth by developing strong long-term relationships with its technology partners and by providing tremendous value to its clients. We pride ourselves on delivering the most innovative technology solutions enabling our customers to lead their market segments and stay ahead of the competition. At Belocal, we pride ourselves in our ability to listen, our attention to detail and our expertise in innovation. Such strengths have enabled us to develop new solutions and services, to suit the changing needs of our clients and acquire new businesses by tailoring all our solutions and services to the specific needs of each client.

View source version on businesswire.com: https://www.businesswire.com/news/home/20200127005311/en/

Contacts

Iguazio Media Contact: Sahar Dolev-Blitental, +972.73.321.0401 press@iguazio.com

Follow this link:

Iguazio Deployed by Payoneer to Prevent Fraud with Real-time Machine Learning - Yahoo Finance

Which-50 and LogMeIn recently surveyed call centre managers and C-Suite executives with responsibility for the customer, asking them to nominate the technologies they believe will be most transformative.

AI & machine learning was nominated by more than three quarters of respondents, making it the top pick.

We asked Ryan Lester, Senior Director, Customer Experience Technologies at LogMeIn, to describe where the short- and longer-term impacts of AI are most likely to be felt, and also to describe the impact on contact centre agents.

Lester told Which-50 that AI is the broader umbrella and machine learning is the algorithms you build to improve the quality of your prediction.

He said brands should be very thoughtful if they are going to do machine learning themselves and invest in machine learning teams. However, he recommended that companies dont do that.

Rather, he said there are plenty of off-the-shelf solutions that are purpose-built for contact centres or for conversion metrics.

He said, You can buy a business application versus buying lets say a machine learning tool or platform.

Lester said that in the immediate term what companies can do to avoid some of the challenges around a bad investment is to use AI as their first round listening mechanism. Brands can leverage a solution built for the contact centre, and it will listen to these customer conversations over phone calls.

Then LogMeIn can see certain intents, Lester said, So Ill say here are intents Im seeing. You can also take large databases. If you have chat records from the last year, you can stick those into AI tools that will start to help you identify intents.

You can take historical data and use it as a place to say, well we should go investigate further here and then start building more purposeful applications around those workflows.

He said companies should build around their existing workflows. They should focus on those workflows today before they invest heavily in either a technology spend or research spend or a headcount spend.

The longer-term impact of machine learning is moving away from inbound response.

Lester said when a customer is contacting a company about a specific problem, the company should operationalise it. That means making it more efficient so making it self-service or reducing delivery costs. They want to align the right resource to the right problem.

Where theres an opportunity longer-term is to think about more of the entire customer lifecycle, he explained.

Lester said AI will help to discover what types of customers brands should be engaging with through leading indicators.

We should start being more proactive about engagement for these types of customers with these types of attributes. If were seeing retention challenges on particular types of customers, we should be offering up those types of offerings to those customers.

He believes many of the conversations are still really about inbound customer service, when in the longer term theres going to be a much bigger opportunity around the entire customer lifecycle.

Saying, for these types of customers we acquired this way, heres how were upselling them, heres how were better retaining them and looking much more at the lifecycle and how AI is helping across that entire lifecycle.

Athina Mallis is the editor of the Which-50 Digital Intelligence Unit of which LogMeIn is a corporate member. Members provide their insights and expertise for the benefit of the Which-5o community. Membership fees apply.

Previous post

Next post

View original post here:

Short- and long-term impacts of machine learning on contact centres - Which-50

The New York Institute of Finance (NYIF) and Google Cloud have launched a new Machine Learning for Trading Specialisation available exclusively on the Coursera platform.

The Specialisation helps learners leverage the latest AI and machine learning techniques for financial trading.

Amid the Fourth Industrial Revolution, nearly 80 per cent of financial institutions cite machine learning as a core component of business strategy and 75 per cent of financial services firms report investing significantly in machine learning. The Machine Learning for Trading Specialisation equips professionals with key technical skills increasingly needed in the financial industry today.

Composed of three courses in financial trading, machine learning, and artificial intelligence, the Specialisation features a blend of theoretical and applied learning. Topics include analysing market data sets, building financial models for quantitative and algorithmic trading, and applying machine learning in quantitative finance.

As we enter an era of unprecedented technological change within our sector, were proud to offer up-skilling opportunities for hedge fund traders and managers, risk analysts, and other financial professionals to remain competitive through Coursera, says Michael Lee, Managing Director of Corporate Development at NYIF. The past ten years have demonstrated the staying power of AI tools in the finance world, further proving the importance for both new and seasoned professionals to hone relevant tech skills.

The Specialisation is particularly suited for hedge fund traders, analysts, day traders, those involved in investment management or portfolio management, and anyone interested in constructing effective trading strategies using machine learning. Prerequisites include basic competency with Python, familiarity with pertinent libraries for machine learning, a background in statistics, and foundational knowledge of financial markets.

Cutting-edge technologies, such as machine and reinforcement learning, have become increasingly commonplace in finance, says Rochana Golani, Director, Google Cloud Learning Services. Were excited for learners on Coursera to explore the potential of machine learning within trading. Looking beyond traditional finance roles, were also excited for the Specialisation to support machine learning professionals seeking to apply their craft to quantitative trading strategies.

Excerpt from:

New York Institute of Finance and Google Cloud launch a Machine Learning for Trading Specialisation on Coursera - HedgeWeek

Workflow-integrated storage supplier Iguazio has received $24m in C-round funding and announced its Data Science Platform. This is deeply integrated into AI and machine learning processes, and accelerates them to real-time speeds through parallel access to multi-protocol views of a single storage silo using data container tech.

The firm said digital payment platform provider Payoneer is using it for proactive fraud prevention with real-time machine learning and predictive analytics.

Yaron Weiss, VP Corporate Security and Global IT Operations (CISO) at Payoneer, said of Iguazios Data Science Platform: Weve tackled one of our most elusive challenges with real-time predictive models, making fraud attacks almost impossible on Payoneer.

He said Payoneer had built a system which adapts to new threats and enables is to prevent fraud with minimum false positives. The systems predictive machine learning models identify suspicious fraud and money laundering patterns continuously.

Weiss said fraud was detected retroactively with offline machine learning models; customers could only block users after damage had already been done. Now it can take the same models and serve them in real time against fresh data.

The Iguazio system uses a low latency serverless framework, a real-time multi-model data engine and a Python eco-system running over Kubernetes. Iguazio claims an estimated 87 per cent of data science models which have shown promise in the lab never make it to production because of difficulties in making them operational and able to scale.

It is based on so-called data containers that store normalised data from multiple sources; incoming stream records, files, binary objects, and table items. The data is indexed, and encoded by a parallel processing engine. Its stored in the most efficient way to reduce data footprint while maximising search and scan performance for each data type.

Data containers are accessed througha V310 API and can be read as any type regardless of how it was ingested. Applications can read, update, search, and manipulate data objects, while the data service ensures data consistency, durability, and availability.

Customers can submit SQL or API queries for file metadata, to identify or manipulate specific objects without long and resource-consuming directory traversals, eliminating any need for separate and non-synchronised file-metadata databases.

So-called API engines engine uses offload techniques for common transactions, analytics queries, real-time streaming, time-series, and machine-learning logic. They accept data and metadata queries, distribute them across all CPUs, and leverage data encoding and indexing schemes to eliminate I/O operations. Iguazio claims this provides magnitudes faster analytics and eliminates network chatter.

The Iguazio software is claimed to be able to accelerate the performance of tools such as Apache Hadoop and Spark by up to 100 times without requiring any software changes.

This DataScience Platform can run on-premises or in the public cloud. The Iguazio website contains much detail about its components and organisation.

Iguazio will use the $24m to fund product innovation and support global expansion into new and existing markets. The round was led by INCapital Ventures, with participation from existing and new investors, including Samsung SDS, Kensington Capital Partners, Plaza Ventures and Silverton Capital Ventures.

See the rest here:

Iguazio pulls in $24m from investors, shows off storage-integrated parallelised, real-time AI/machine learning workflows - Blocks and Files

A few years ago, I wondered how edge data would ever be useful given the enormous cost of transmitting all the data to either the centralized data center or some variant of cloud infrastructure. (It is said that 5G will solve that problem).

Consider, for example, applications of vast sensor networks that stream a great deal of data at small intervals. Vehicles on the move are a good example.

There is telemetry from cameras, radar, sonar, GPS and LIDAR, the latter about 70MB/sec. This could quickly amount to four terabytes per day (per vehicle). How much of this data needs to be retained? Answers I heard a few years ago were along two lines:

My counterarguments at the time were:

Introducing TensorFlow federated, via The TensorFlow Blog:

This centralized approach can be problematic if the data is sensitive or expensive to centralize. Wouldn't it be better if we could run the data analysis and machine learning right on the devices where that data is generated, and still be able to aggregate together what's been learned?

Since I looked at this a few years ago, the distinction between an edge device and a sensor has more or less disappeared. Sensors can transmit via wifi (though there is an issue of battery life, and if they're remote, that's a problem); the definition of the edge has widened quite a bit.

Decentralized data collection and processing have become more powerful and able to do an impressive amount of computing. The case is point in Intel's Introducing the Intel Neural Compute Stick 2 computer vision and deep learning accelerator powered by the Intel Movidius Myriad X VPU, that can stick into a Pi for less than $70.00.

But for truly distributed processing, the Apple A13 chipset in the iPhone 11 has a few features that boggle the mind: From Inside Apple's A13 Bionic system-on-chip Neural Engine, a custom block of silicon separate from the CPU and GPU, focused on accelerating Machine Learning computations. The CPU has a set of "machine learning accelerators" that perform matrix multiplication operations up to six times faster than the CPU alone. It's not clear how exactly this hardware is accessed, but for tasks like machine learning (ML) that use lots of matrix operations, the CPU is a powerhouse. Note that this matrix multiplication hardware is part of the CPU cores and separate from the Neural Engine hardware.

This should beg the question, "Why would a smartphone have neural net and machine learning capabilities, and does that have anything to do with the data transmission problem for the edge?" A few years ago, I thought the idea wasn't feasible, but the capability of distributed devices has accelerated. How far-fetched is this?

Let's roll the clock back thirty years. The finance department of a large diversified organization would prepare in the fall a package of spreadsheets for every part of the organization that had budget authority. The sheets would start with low-level detail, official assumptions, etc. until they all rolled up to a small number of summary sheets that were submitted headquarters. This was a terrible, cumbersome way of doing things, but it does, in a way, presage the concept of federated learning.

Another idea that vanished is Push Technology that shared the same network load as centralizing sensor data, just in the opposite direction. About twenty-five years, when everyone had a networked PC on their desk, the PointCast Network used push technology. Still, it did not perform as well as expected, often believed to be because its traffic burdened corporate networks with excessive bandwidth use, and was banned in many places. If Federated Learning works, those problems have to be addressed

Though this estimate changes every day, there are 3 billion smartphones in the world and 7 billion connected devices.You can almost hear the buzz in the air of all of that data that is always flying around. The canonical image of ML is that all of that data needs to find a home somewhere so that algorithms can crunch through it to yield insights. There are a few problems with this, especially if the data is coming from personal devices, such as smartphones, Fitbit's, even smart homes.

Moving highly personal data across the network raises privacy issues. It is also costly to centralize this data at scale. Storage in the cloud is asymptotically approaching zero in cost, but the transmission costs are not. That includes both local WiFi from the devices (or even cellular) and the long-distance transmission from the local collectors to the central repository. This s all very expensive at this scale.

Suppose, large-scale AI training could be done on each device, bringing the algorithm to the data, rather than vice-versa? It would be possible for each device to contribute to a broader application while not having to send their data over the network. This idea has become respectable enough that it has a name - Federated Learning.

Jumping ahead, there is no controversy that training a network without compromising device performance and user experience, or compressing a model and resorting to a lower accuracy are not alternatives. In Federated Learning: The Future of Distributed Machine Learning:

To train a machine learning model, traditional machine learning adopts a centralized approach that requires the training data to be aggregated on a single machine or in a datacenter. This is practically what giant AI companies such as Google, Facebook, and Amazon have been doing over the years. This centralized training approach, however, is privacy-intrusive, especially for mobile phone usersTo train or obtain a better machine learning model under such a centralized training approach, mobile phone users have to trade their privacy by sending their personal data stored inside phones to the clouds owned by the AI companies.

The federated learning approach decentralizes training across mobile phones dispersed across geography. The presumption is that they collaboratively develop machine learning while keeping their personal data on their phones. For example, building a general-purpose recommendation engine for music listeners. While the personal data and personal information are retained on the phone, I am not at all comfortable that data contained in the result sent to the collector cannot be reverse-engineered - and I havent heard a convincing argument to the contrary.

Here is how it works. A computing group, for example, is a collection of mobile devices that have opted to be part of a large scale AI program. The device is "pushed" a model and executes it locally and learns as the model processes the data. There are some alternatives to this. Homogeneous models imply that every device is working with the same schema of data. Alternatively, there are heterogeneous models where harmonization of the data happens in the cloud.

Here are some questions in my mind.

Here is the fuzzy part: federated learning sends the results of the learning as well as some operational detail such as model parameters and corresponding weights back to the cloud. How does it do that and preserve your privacy and not clog up your network? The answer is that the results are a fraction of the data, and since the data itself is not more than a few Gb, that seems plausible. The results sent to the cloud can be encrypted with, for example, homomorphic encryption (HE). An alternative is to send the data as a tensor, which is not encrypted because it is not understandable by anything but the algorithm. The update is then aggregated with other user updates to improve the shared model. Most importantly, all the training data remains on the user's devices.

In CDO Review, The Future of AI. May Be In Federated Learning:

Federated Learning allows for faster deployment and testing of smarter models, lower latency, and less power consumption, all while ensuring privacy. Also, in addition to providing an update to the shared model, the improved (local) model on your phone can be used immediately, powering experiences personalized by the way you use your phone.

There is a lot more to say about this. The privacy claims are a little hard to believe. When an algorithm is pushed to your phone, it is easy to imagine how this can backfire. Even the tensor representation can create a problem. Indirect reference to real data may be secure, but patterns across an extensive collection can surely emerge.

Originally posted here:

Federated machine learning is coming - here's the questions we should be asking - Diginomica

Despite being one of the richest countries in the world, in Qatar, digital maps are lagging behind. While the country is adding new roads and constantly improving old ones in preparation for the 2022 FIFA World Cup, Qatar isn't a high priority for the companies that actually build out maps, like Google.

"While visiting Qatar, weve had experiences where our Uber driver cant figure out how to get where hes going, because the map is so off," Sam Madden, a professor at MIT's Department of Electrical Engineering and Computer Science, said in a prepared statement. "If navigation apps dont have the right information, for things such as lane merging, this could be frustrating or worse."

Madden's solution? Quit waiting around for Google and feed machine learning models a whole buffet of satellite images. It's faster, cheaper, and way easier to obtain satellite images than it is for a tech company to drive around grabbing street-view photos. The only problem: Roads can be occluded by buildings, trees, or even street signs.

So Madden, along with a team composed of computer scientists from MIT and the Qatar Computing Research Institute, came up with RoadTagger, a new piece of software that can use neural networks to automatically predict what roads look like behind obstructions. It's able to guess how many lanes a given road has and whether it's a highway or residential road.

RoadTagger uses a combination of two kinds of neural nets: a convolutional neural network (CNN), which is mostly used in image processing, and a graph neural network (GNN), which helps to model relationships and is useful with social networks. This system is what the researchers call "end-to-end," meaning it's only fed raw data and there's no human intervention.

First, raw satellite images of the roads in question are input to the convolutional neural network. Then, the graph neural network divides up the roadway into 20-meter sections called "tiles." The CNN pulls out relevant road features from each tile and then shares that data with the other nearby tiles. That way, information about the road is sent to each tile. If one of these is covered up by an obstruction, then, RoadTagger can look to the other tiles to predict what's included in the one that's obfuscated.

Parts of the roadway may only have two lanes in a given tile. While a human can easily tell that a four-lane road, shrouded by trees, may be blocked from view, a computer normally couldn't make such an assumption. RoadTagger creates a more human-like intuition in a machine learning model, the research team says.

Neural Net Can Solve Calculus Equation in 1 Second

Neural Networks Now 90% Smaller, but Just as Smart

Neural Networks Can't Write Harry Potter Books

"Humans can use information from adjacent tiles to guess the number of lanes in the occluded tiles, but networks cant do that," Madden said. "Our approach tries to mimic the natural behavior of humans ... to make better predictions."

The results are impressive. In testing out RoadTagger on occluded roads in 20 U.S. cities, the model correctly counted the number of lanes 77 percent of the time and inferred the correct road types 93 percent of the time. In the future, the team hopes to include other new features, like the ability to identify parking spots and bike lanes.

Here is the original post:

How Machine Learning Will Lead to Better Maps - Popular Mechanics

(Robert Lucian Crusitu/Shutterstock)

Theres no shortage of resources and tools for developing machine learning algorithms. But when it comes to putting those algorithms into production for inference, outside of AWSs popular SageMaker, theres not a lot to choose from. Now a startup called Cortex Labs is looking to seize the opportunity with an open source tool designed to take the mystery and hassle out of productionalizing machine learning models.

Infrastructure is almost an afterthought in data science today, according to Cortex Labs co-founder and CEO Omer Spillinger. A ton of energy is going into choosing how to attack problems with data why, use machine learning of course! But when it comes to actually deploying those machine learning models into the real world, its relatively quiet.

We realized there are two really different worlds to machine learning engineering, Spillinger says. Theres the theoretical data science side, where people talk about neural networks and hidden layers and back propagation and PyTorch and Tensorflow. And then you have the actual system side of things, which is Kubernetes and Docker and Nvidia and running on GPUs and dealing with S3 and different AWS services.

Both sides of the data science coin are important to building useful systems, Spillinger says, but its the development side that gets most of the glory. AWS has captured a good chunk of the market with SageMaker, which the company launched in 2017 and which has been adopted by tens of thousands of customers. But aside from just a handful of vendors working in the area, such as Algorithmia, the general data-building public has been forced to go it alone when it comes to inference.

A few years removed from UC Berkeleys computer science program and eager to move on from their tech jobs, Spillinger and his co-founders were itching to build something good. So when it came to deciding what to do, Spillinger and his co-founders decided to stick with what they knew, which was working with systems.

(bluebay/Shutterstock.com)

We thought that we could try and tackle everything, he says. We realized were probably never going to be that good at the data science side, but we know a good amount about the infrastructure side, so we can help people who actually know how to build models get them into their stack much faster.

Cortex Labs software begins where the development cycle leaves off. Once a model has been created and trained on the latest data, then Cortex Labs steps in to handle the deployment into customers AWS accounts using its Kubernetes engine (AWS is the only supported cloud at this time; on-prem inference clusters are not supported).

Our starting point is a trained model, Spillinger says. You point us at a model, and we basically convert it into a Web API. We handle all the productionalization challenges around it.

That could be shifting inference workloads from CPUs to GPUs in the AWS cloud, or vice versa. It could be we automatically spinning up more AWS servers under the hood when calls to the ML inference service are high, and spinning down the servers when that demand starts to drop. On top of its build-in AWS cost-optimization capabilities, the Cortex Labs software logs and monitors all activities, which is a requirement in todays security- and regulatory-conscious climate.

Cortex Labs is a tool for scaling real-time inference, Spillinger says. Its all about scaling the infrastructure under the hood.

Cortex Labs delivers a command line interface (CLI) for managing deployments of machine learning models on AWS

We dont help at all with the data science, Spillinger says. We expect our audience to be a lot better than us at understanding the algorithms and understanding how to build interesting models and understanding how they affect and impact their products. But we dont expect them to understand Kubernetes or Docker or Nvidia drivers or any of that. Thats what we view as our job.

The software works with a range of frameworks, including TensorFlow, PyTorch, scikit-learn, and XGBoost. The company is open to supporting more. Theres going to be lots of frameworks that data scientists will use, so we try to support as many of them as we can, Spillinger says.

Cortex Labs software knows how to take advantage of EC2 spot instances, and integrates with AWS services like Elastic Kubernetes Service (EKS), Elastic Container Service (ECS), Lambda, and Fargate. The Kubernetes management alone may be worth the price of admission.

You can think about it as a Kubernetes thats been massaged for the data science use case, Spillinger says. Theres some similarities to Kubernetes in the usage. But its a much higher level of abstraction because were able to make a lot of assumptions about the use case.

Theres a lack of publicly available tools for productionalizing machine learning models, but thats not to say that they dont exist. The tech giants, in particular, have been building their own platforms for doing just this. Airbnb, for instance, has its BigHead offering, while Uber has talked about its system, called Michelangelo.

But the rest of the industry doesnt have these machine learning infrastructure teams, so we decided wed basically try to be that team for everybody else, Spillinger says.

Cortex Labs software is distributed under an open source license and is available for download from its GitHub Web page. Making the software open source is critical, Spillinger says, because of the need for standards in this area. There are proprietary offerings in this arena, but they dont have a chance of becoming the standard, whereas Cortex Labs does.

We think that if its not open source, its going to be a lot more difficult for it to become a standard way of doing things, Spillinger says.

Cortex Labs isnt the only company talking about the need for standards in the machine learning lifecycle. Last month, Cloudera announced its intention to push for standards in machine learning operations, or MLOps. Anaconda, which develops a data science platform, also is backing

Eventually, the Oakland, California-based company plans to develop a managed service offering based on its software, Spillinger says. But for now, the company is eager to get the tool into the hands of as many data scientists and machine learning engineers as it can.

Related Items:

Its Time for MLOps Standards, Cloudera Says

Machine Learning Hits a Scaling Bump

Inference Emerges As Next AI Challenge

Read the rest here:

An Open Source Alternative to AWS SageMaker - Datanami