HP TECH TAKES /...
Exploring today's technology for tomorrow's possibilities
Subscribe
Innovation in Tech: Exploring the Impact of AI and Machine Learning in the UK
Stephen Mash
|
June 12, 2023
Machine learning is the ability of a computer to learn how to operate without explicit programming of how to perform that operation. The computer effectively uncovers the best way to achieve the results rather than following the same steps a human operator would follow.
Machine learning is a part of the broad field of artificial intelligence (AI), the ability of technology to exhibit intelligent behaviour. While the media like to focus on killer robots when discussing AI, the reality is often far more benign. AI in UK tech is gaining prominence with various potential applications; examples include financial management, intelligent crop management, cyber security protection and automated customer services.
One of the most prevalent AI and machine learning applications are product and service support chatrooms, where the website visitor interacts with a computer that attempts to understand their problem and provide a solution. Replacing call centre support staff with an AI-based computer offers enormous savings for businesses as long as the system works and customer service levels remain sufficiently high.
The impact and benefits of machine learning
One of the most promising applications of AI and machine learning-based systems is performing complex tasks where systems cannot be programmed using formally defined processes. This may be due to time constraints or the inability to define the requirements deterministically. A good example is diagnostic image recognition. A highly experienced and knowledgeable expert can competently identify signs of medical conditions from X-rays. However, formally defining the vast range of indicators they seek in terms a computer can emulate is near impossible.
AI-based solutions can learn to achieve image recognition results quicker by applying machine learning techniques to training data. These images have been previously analysed by a human expert and annotated to indicate if any conditions are present and, if yes, which. The computer can then analyse the training data and recognise patterns that form its internally trained model. This trained model can ultimately be applied to real-world unanalysed images and generate accurate condition identification.
The advantage of machine learning is its ability to recognise non-obvious patterns and be unconstrained by human biases and experiences. The system has effectively programmed itself based on its own learned understanding.
The downside to the machine learning process is that it requires enormous volumes of training data to be collated and prepared. The greater the bulk and quality of training data volume, the more accurate the learned patterns and the better the generated outcomes.
AI and machine learning-based systems are ideal for descriptive, predictive, and prescriptive operations. Descriptive functions use data to describe why an event occurred based on known experience. Predictive functions use data to predict what events may happen in the future based on behaviour patterns. Prescriptive functions use data to define what actions should be taken based on previous experience. Machine learning applications will encompass one or more of these types of operations.
The main benefit of machine learning in UK tech is that once trained and its accuracy validated, the system can continuously run, processing information at great speed while delivering consistent results. Ideal for tasks that would quickly exhaust a human operator, such as scanning CT scans images for injuries or aerial crop photographs for evidence of pests and diseases.
HP Enterprise machine learning development systems
HP and AI innovation are synonymous. HP Enterprise is at the forefront of support for AI projects with its HPE Machine Learning Development Environment. This service offers developers and scientists the fastest and easiest way to build machine learning models by providing a solution that trains models faster and creates more accurate models. Removing the complexity and cost associated with infrastructure coding and machine learning model development allows a greater focus on innovation, simplifying provisioning, deployment and management while accelerating time-to-production. Learn more about HPE Machine Learning Development Environment and how it can benefit machine learning development.
HP envisions using workstations for AI and machine learning development as the gateway to automating processes such as facial identification, sentiment analysis, fraud detection and predictive analytics. To support this vision, HP has developed its Z Workstations as industry-leading machine learning solutions that offer developers more control, better performance and added security over alternative options such as cloud-based solutions. This product range includes the HP Z8, the world’s most powerful workstation for machine learning development.
To support the hardware products, the HP ML Developers Portal offers development support for machine learning models, including tools and documentation necessary to install a high-performance, validated machine learning environment on one of our HP Z Workstations. Access the HP Developers Portal and join the HP Workstations Data Science Group to learn more about HP and machine learning.
The built-in security features of HP Z Workstations include self-healing, hardware-enforced, and manageable security solutions making these the world’s most secure desktop workstations. This protection for AI and machine learning developers helps secure valuable intellectual property and private data against evolving malware threats.
HP also offers its Device as a Service (DaaS) to AI and machine learning solutions. HP DaaS is a modern, multi-OS service model that simplifies how commercial organisations provide their users with the correct hardware, support and lifecycle services. This service ensures businesses can optimise end-user productivity, IT efficiency and cost predictability.
The HP Z Series Products
The machine learning process requires vast amounts of data computational resources to be effective, requiring computers with large amounts of fast storage and computational power. For example, the design of the HP Z8 G4 workstation focuses on hosting the large amounts of fast storage and computational power needed for complex machine-learning training. It is also designed for continuous operation to undertake training sessions that can last many days.
The HP Z8 Tower G4 Workstation offers exceptional processing performance for powering through advanced machine learning algorithms, big data processing, and complex simulations. This workstation comes with the Intel® Xeon® Gold 5220 processor with 18 cores running at speeds up to 3.9 GHz using the Intel turbo boost technology. However, it will support running two next-generation Intel Xeon processors with 56 processor cores and up to 3TB of high-speed memory if greater performance is necessary.
Separate graphic process units can also perform mathematical operations, taking some of the load off the main processor and significantly speeding up the calculations involved in the training process. HP computers use exceptional NVIDIA® graphics cards to offer impressive performance throughput that complex machine learning requires.
For less demanding machine learning applications, the HP Z4 Tower G4 Workstation is perfect with its 10th Generation Intel Core™ i9 processor with 12 cores running at speeds up to 4.6 GHz using Intel’s turbo boost technology. There’s also the option of upgrading to the more capable Intel Xeon W-2245 processor running 8 cores at speeds up to 4.5 GHz.
The HP Z2 Mini G9 Workstation offers users the power to run machine learning applications in a mini workstation product with a small physical footprint for spatially challenging working environments. This pocket-sized processing powerhouse comes with either Intel’s 12th Generation Core i7 12700 processor or the more capable Core i9 12900K option. The Core i7 option offers 12 cores running at boosted speeds of up to 4.9 GHz, while the Core i9 option offers 16 cores running at up to 5.2 GHz. There is also a choice of NVIDIA graphics cards to supplement the processing capabilities of this mini workstation.
The HP ZBook Fury G9 16-inch Mobile Workstation offers machine learning processing power in a portable laptop where mobility is required. The most capable option includes the Intel Core i9 12950HX 12th Generation processor running 16 cores at up to 5.0 GHz using boost technology. The accomplished NVIDIA RTX A3000 graphic cards provide graphical processing performance, while a 16-inch widescreen ultra extended graphics array (WUXGA) display with 1920 x 1200 resolution offers incredible visualisation of the results.
The HP ZBook Studio G9 16-inch Mobile Workstation offers a great starting point for running machine learning applications with its affordable processing capabilities in a flexible multi-purpose business laptop. The 12th Generation Intel Core i7 12800H processor, supported by an NVIDIA GeForce RTX™ 3070 graphics card, is ideal for simultaneously running processor-heavy applications.
To discover more about our Z family of workstations, check out the ZCentral Workstations Information on our website.
Summary
Artificial intelligence is in the news with the publication of the government’s 10-year plan to make the UK a global AI superpower by implementing the National AI Strategy. They aim to use AI to increase resilience, productivity, growth and innovation.
This strategy, in part, leverages the role of AI at the heart of business’s digital transformations and the implementation of the fourth industrial revolution, focusing on unleashing the value within business information. However, technology still has a long way to go before becoming a more widespread fixture. Future developments include combining complex machine learning techniques with Bayesian reasoning processes, potentially creating more accurate trained systems from less information.
HP is at the forefront of AI and machine learning development in terms of its computer products and its development support services. The Z family of desktop and mobile workstations handle the computational processing challenges of machine learning. Alongside this hardware, our machine-learning development environment enables research to progress faster and more efficiently.
