Organisations are looking to AI models to bring out digital transformation in business. But, understanding which kind of models are most suited to the business needs is crucial

As enterprises are discovering the benefits of artificial intelligence (AI), they realise the journey to AI is long and bumpy. Many CIOs (chief information officers) want AI to quickly transform their business without identifying which processes will perform better with AI. In an ideal world, one can pick any process, infuse it with AI and then discover the pros and cons during the journey. The best way to learn is on projects rather than researching through theoretical case studies.

But there needs to be some order to the madness. Can we generalise some patterns that could make it easy for business owners to apply AI? Let’s discuss some scalable, enterprise-relevant AI patterns.

Discover new processes: This is about finding new opportunities afforded by AI. Consider the example of a defect in a machine, degrading over time. An experienced mechanical engineer can deduct the condition of the machine from the sound generated. What if there was a process wherein the mechanical engineer documents what he ‘hears’ and how he maintains the machine? This is where an acoustic AI model can be created, which can analyse sound samples of the machine to predict failures. It’s common sense for an engineer that a noisy machine is the first sign of mechanical failure; shouldn’t this important data can be put to value?

Imagine driving down a highway and an alert pops up on the dashboard saying, “Possible less lubricant”. It confirms the driver’s gut feeling that there’s something wrong in the car.

Most of the acoustic models today use humans to classify the data fed into the model; over time, the model learns to classify on its own. For example, we need to gather at least 10,000 sound clips of failed and normal ball bearings to classify the anomalous sounds and detect the issue. Not an easy task, but it can tremendously help in predicting failures in mines, underground subways, nuclear plants and highly critical sites unapproachable by humans.

Reinvigorate old processes: This pattern improves existing processes by introducing AI. For instance, almost every organisation collects data from the employees’ badges, which provides information on access control and employee movement. Reinvigorating this process by adding occupancy sensors and then adding AI will help derive deeper insights such as the number of people per floor. This data can be fed into an AI model to predict the occupancy rate and help organisations reduce the cost associated with each desk and decide how much office space should be leased or vacated. The ability to predict churn of clients, machine failure, energy usage etc are all examples of how old processes can be reinvigorated with new AI models.

Unlock data: Organisations can derive value from their data by applying AI. For example, machine learning algorithms can be used to detect fraud in financial transactions or even an asset defect, which would otherwise go unnoticed by humans. One Machine learning model can be fed time-series data to discover patterns of anomalies, while another can be fed asset manuals to find contextual text of the faults. One of the widely used examples of applying AI to businesses is handling unstructured data in the form of texts, videos, and tweets. Several organisations across industries have benefitted from this pattern, including telcos with millions of call records, banks with loan records, manufacturing units with work orders etc.

Opening new channels: This is another area where we have seen several businesses apply AI successfully. This essentially means starting a new channel of interaction with customers or employees using AI-based virtual assistants with natural language processing technologies. Unlike the dated IVR system, this new channel is helping organisations reach their clients and service them in unique ways.

We can pick any AI process and it’s sure to fall in one of the four patterns mentioned above. What is needed is the right understanding of which process to choose and then applying the right AI methodology to solve the business problem. Once the process is chosen, along with the the right algorithm and data quality, one also needs to check for bias in the model. Explanation of why a certain recommendation is the most right one needs to be included too.

The post How to choose the right pattern to infuse AI into your business appeared first on Artificial Intelligence.

An object in motion looks fundamentally different from an object at rest — especially to a computer. To get a better idea of this concept, let’s imagine a film strip of a sprinter running: The person and pose in one frame look drastically different from the next frame, right?

Making sense of dynamic objects is taking on new importance as cities begin incorporating IoT devices like smart cameras to streamline municipal life. The town of Yuma, Arizona, is a great example of this. The city recently installed cameras on streetlights that can detect when cars, bicycles, and pedestrians travel through intersections, and it uses that data to optimise signal switching.

Athena Security is pioneering another interesting application of moving-video analysis: The company sells software that uses artificial intelligence to detect when people are fighting, fleeing, or lurking to determine whether crimes are being committed (or are imminent). Unsurprisingly, everyone from municipal police departments to Fortune 500 companies is interested in this AI application.

The applications are endless for IoT devices like smart cameras that analyse moving video. Fortunately, this technology has now reached a point where almost anything is possible.

• Vulnerabilities in smart IP cameras expose users to privacy, security risks

Solving mysteries in moving video

Using computers to analyse video isn’t exactly a new concept. However, there’s one problem hampering the development of video analysis: Moving video is full of dynamic variables that can confuse even the smartest computers.

Objects look completely different in low light compared to bright light, for instance, which can lead to false analyses. Perspective offers up another challenge: Think about how different a car looks when it’s traveling parallel and then perpendicular to a relative point.

Other issues that might be confusing for a machine’s analysis of video include moving shadows, complex backgrounds, obscured objects, unexpected movements, and a camera’s technical limitations. For all these reasons, moving-video analysis has historically had a lot of potential — but not too many practical applications.

That’s all changing with advances in deep learning systems, which are often referred to as neural networks. Today, computing has advanced to a point where systems can learn from past data to get better at understanding future data.

The ability to learn and adapt is crucial for computers that need to make sense of the ever-changing data coming from moving video — and different combinations of neural networks could provide a solution. With convolutional neural networks, for example, computers model space in three dimensions to better predict the trajectory of objects within that space.

Deep neural networks can help cancel out background images so cameras can focus explicitly on moving objects. There are also recurrent neural networks that excel at pattern recognition. Each of these networks has strengths and weaknesses, but using them in the right combination makes moving-video analysis highly accurate in almost any setting.

• Hackers could spy on you through your smart camera

Connectivity and the future of smart cameras

My company recently worked on a project that demonstrates how far connected devices like smart cameras have come, as well as the challenges they still face. For this particular project, a client in Israel asked us to develop a program to detect kicking motions in live soccer matches televised at 20 frames per second.

From the start, the project endured two obstacles: First, our team had to distinguish between an actual “kick” and a swinging leg motion that looked quite similar. Second, we needed to do that at 20 frames per second. That’s a higher resolution than most surveillance footage, and it’s packed with much more data to analyse.

Initially, we tried developing an algorithm that would create two “bounding boxes” around both a player’s foot and the soccer ball, and then register when those two boxes met. In practice, however, detecting kicks became extremely inaccurate when players were clumped together (and we know that happens a lot in soccer).

The solution? Tweak the deep learning element. We adjusted how the underlying neural networks were configured so we could accelerate object detection. Then, we created a data set using 500 frames taken from 20 seconds of a soccer match. Our team manually annotated this data to identify kicks and “non-kicks,” and we used it to “teach” our algorithms to make that distinction.

Our program eventually identified 58 per cent of real kicks; improving the numbers was possible through feeding the program data from more matches and different sets of players.

That’s because it proved that, with the right configurations, deep learning can make sense of all the complexities within the moving video within the connected device. While achieving these ends might take a ton of reference data, the technology has proved its usefulness and has finally made moving-video analysis a reality.

This kind of technology can be applied in many areas, from IoT surveillance systems to self-driving cars. And if one thing is certain, there’s not much further to go before moving-video analysis begins transforming our lives.

The post Unlocking the potential of smart cameras with deep learning appeared first on Artificial Intelligence.

The Apache Software Foundation (ASF) recently announced that SINGA, a framework for distributed deep-learning, has graduated to top-level project (TLP) status, signifying the project’s maturity and stability. SINGA has already been adopted by companies in several sectors, including banking and healthcare.

Originally developed at the National University of Singapore, SINGA joined ASF’s incubator in March 2015. SINGA provides a framework for distributing the work of training deep-learning models across a cluster of machines, in order to reduce the time needed to train the model. In addition to its use as a platform for academic research, SINGA has been used in commercial applications by Citigroup and CBRE, as well as in several health-care applications, including an app to aid patients with pre-diabetes.

The success of deep-learning models has been driven by the use of very large datasets, such as ImageNet with hundreds of thousands of images, and complex models with millions of parameters. Google’s BERT natural-language model contains 300 million parameters and is trained on nearly 3 billion words. However, this training often requires hours, if not days, to complete. To speed up this process, researchers have turned to parallel computing, which distributes the work across a cluster of machines. According to Professor Beng Chin Ooi, leader of the research group that developed SINGA:

It is essential to scale deep learning via distributed computing as…deep learning models are typically large and trained over big datasets, which may take hundreds of days using a single GPU.

There are two broad parallelism strategies for distributed deep-learning: data parallelism, where multiple machines work on different subsets of the input data, and model parallelism, where multiple machines train different sections of the neural-network model. SINGA supports both of these strategies, as well as a combination of the two. These strategies do introduce some communication and synchronization overhead, required to coordinate the work among the machines in the cluster. SINGA implements several optimizations to minimize this overhead.

Acceptance as a top-level project means that SINGA has passed several milestones related to software quality and community, which in theory makes the software more attractive as a solution. However, one possible barrier to adoption is that instead of building upon an existing API for modeling neural networks, such as Keras, SINGA’s designers chose to implement their own. By contrast, the Horovod framework open-sourced by Uber allows developers to port existing models written for the two most popular deep-learning frameworks, TensorFlow and PyTorch. PyTorch in particular is the framework used in a majority of recent research papers.

ASF has several other top-level distributed-data processing projects that support machine-learning, including Spark and Ignite. Unlike these, SINGA is designed specifically for deep-learning’s large models. ASF is also home to MXNet, a deep-learning framework similar to TensorFlow and PyTorch, which is still in incubator status. AWS touted MXNet as its framework of choice in late 2016, but MXNet still hasn’t achieved widespread popularity, hovering at just under 2% in KDNugget’s polls.

Apache SINGA version 2.0 was released in April, 2019. The source code is available on GitHub, and a list of open issues can be tracked in SINGA’s Jira project. According to ASF, upcoming features include “SINGA-lite for deep learning on edge devices with 5G, and SINGA-easy for making AI usable by domain experts (without deep AI background).

The post Deep-Learning Framework SINGA Graduates to Top-Level Apache Project appeared first on Artificial Intelligence.