After having learned about the basics of AI, how did AI evolve through the last decades, and how do different countries apply AI solutions in practice you are invited to get familiar with what exactly AI, ML, DL and RPA mean.
AI general applications
Definitions of AI, ML and DL, RPA
Artificial Intelligence (AI)
Watch the below video and think of similar examples of AI real world use cases you have heard of, read about or you know of:
Heuristic video
(
As explained also in the above video, at its simplest form, artificial intelligence combines computer science and robust datasets, to enable problem-solving, decision making, etc.
Artificial intelligence (AI) is the ability of a computer or a robot controlled by a computer to do tasks that are usually done by humans because they require human intelligence and discernment.
AI offers a wide range of tools helping to evaluate and produce better information integration, and data analysis to improve decision-making. This can be done by leveraging data collected from various sources, across different platforms and spectrums. This will allow AI to offer better capability of tremendous sophistication and analytics. The key to better analysis of sophisticated trends and interconnections is through taking advantage of powerful AI capabilities. By having access to enormous amounts of data from different industries and ecosystems like finance, healthcare, smart cities, IoT, etc. AI will be able to find value in areas across various unrelated fields, and modify its capability to produce superior results in their field of application.
These disciplines are comprised of AI algorithms, which seek to create expert systems that make predictions or classifications based on input data.
A prediction is anticipating an occurrence or event that has not yet happened based on information that leads to it happening.
Classification is the process of putting equals with equals, forming groups with homogeneous members (e.g. distinguishing pictures of dogs and cats). Classification is done on the basis of certain characteristics based on differential data of the elements to be classified (e.g. the shape of the nose or eyes of cats).
In general, AI systems work by ingesting large amounts of labeled training data, analyzing the data for correlations and patterns, and using these patterns to make predictions about future states.
In this way, for instance, a chatbot that is fed with examples of text chats can learn to produce lifelike exchanges with people, or an image recognition tool can learn to identify and describe objects in images by reviewing millions of examples.
AI will also help predict trends in the market and generate products that are in line with the upcoming trends with much better accuracy. It can also better manage logistics and the supply chain.
Artificial Intelligence is the broader umbrella under which Machine Learning and Deep Learning come. And as the diagram shows below Deep Learning is a subset of Machine Learning. All three of them AI, machine learning and deep learning are just the subsets of each other. So let us move on and understand how exactly they are different from each other.
(source: edureka.co)
Machine Learning (ML)
After watching the Youtube video think of similar AI use cases from your daily life.
)
OTHER ML examples from the PERFORM AI videos:
- Cape AI
- ARUTEC
It means the ability of the AI based tool to learn. ML is defined by Computer Scientist and machine learning pioneer Tom M. Mitchell: “Machine learning is the study of computer algorithms that allow computer programs to automatically improve through experience.”
ML is one of the ways we expect to achieve AI. Machine learning relies on working with small to large datasets by examining and comparing the data to find common patterns and explore nuances.
The function of a machine learning system can be
- descriptive, meaning that the system uses the data to explain what happened (e.g. Social media usage and engagement data such as Instagram or Facebook likes);
- predictive, meaning the system uses the data to predict what will happen (e.g. managing performance expectations and avoiding risks); or
- prescriptive, meaning the system will use the data to make suggestions about what action to take (e.g. investment decisions).
We can mention different types of ML, supervised ML, unsupervised ML, Reinforcement ML. Whereas Federated ML is a way to produce machine learning from several servers working with distributed data sets, without sharing data. It is also called collaborative learning.
Deep Learning (DL)
In fact, all deep learning is machine learning, but not all machine learning is deep learning.
Deep learning is a subset of machine learning, which is essentially a neural network with many layers. These neural networks attempt to simulate the behavior of the human brain—albeit far from matching its ability—allowing it to “learn” from large amounts of data. While a neural network with a single layer can still make approximate predictions, additional hidden layers can help to optimize and refine for accuracy.Deep learning drives many artificial intelligence (AI) applications and services that improve automation, performing analytical and physical tasks without human intervention. Deep learning technology lies behind everyday products and services (such as digital assistants, voice-enabled TV remotes, and credit card fraud detection) as well as emerging technologies (such as self-driving cars).
Further, medical diagnostics-related examples include
monitoring acute macular degeneration,
quantifying fatigue attributable to multiple sclerosis, and
measuring the behavioral manifestations of genetic autism disorders in preclinical models.
Deep learning began to perform tasks that were impossible to do with classic rule-based programming. Fields such as speech and face recognition, image classification, and natural language processing, which were at early stages, suddenly took great leaps.
quantifying fatigue attributable to multiple sclerosis, and
measuring the behavioral manifestations of genetic autism disorders in preclinical models.
Deep learning began to perform tasks that were impossible to do with classic rule-based programming. Fields such as speech and face recognition, image classification, and natural language processing, which were at early stages, suddenly took great leaps.
Robotic Process Automation (RPA)
RPA technology is changing how the world gets work done. Software robots—instead of people—do repetitive and lower-value work, like logging into applications and systems, moving files and folders, extracting, copying, and inserting data, filling in forms, and completing routine analyses and reports. Advanced robots can even perform cognitive processes, like interpreting text, engaging in chats and conversations, understanding unstructured data, and applying advanced machine learning models to make complex decisions. When robots do these types of repetitive, high-volume tasks, humans are freed to focus on the things they do best and enjoy more: innovating, collaborating, creating, and interacting with customers.
Enterprises get a boost too:
efficiency, and
resilience.
It is no wonder that RPA is rewriting the story of work.
To demonstrate just how widespread RPA’s usefulness is (and why so many are investing in it), we’ve compiled a list of real-life examples of RPA helping companies reduce cost, save time, stay competitive and do better work.
Call centre operations
Many of the customer requests received by call centres can be supported with RPA technology; common customer queries and solutions can be provided to agents via a dashboard. When an issue gets escalated to human customer service agents, RPA can help consolidate all the information about a customer on a single screen, so the agents have all the information they need from multiple systems to provide exemplary service.
Credit card applications
Today, bots are behind the scenes processing the majority of credit card applications. They can be programmed to easily handle all aspects of the process from gathering information and documents, doing
credit and background cheques, and ultimately deciding if the applicant is worthy of receiving credit and issuing the actual card.
Help desk
As the first line to address user’s technical problems, RPA can help diminish the workload of the human help desk by taking care of straightforward, repetitive issues. These level-one tech support issues are simple but time-consuming. In addition, regular diagnostic tests of a company’s computer systems completed by bots will help the human IT staff stay ahead of issues.
Further resources: Videos and/or useful links
● Introduction to AI:
● Documentary movie about AI:
References
● https://www.techtarget.com/searchenterpriseai/definition/AI-Artificial-Intelligence
● Hatamleh, Omar; Tilesch, George. BetweenBrains: Taking back our AI Future. Kindle Edition, https://betweenbrains.ai/
● Machine Learning definition, 2020, Expert AI, https://www.expert.ai/blog/machine-learning-definition/
● Machine Learning, 2017 SAS Institute Inc., sas.com/content/dam/SAS/documents/marketing-whitepapers-ebooks/sas-whitepapers/en/machine-learning-primer-108796.pdf
● Artificial Intelligence: The Ambiguous Labor Market Impact of Automating Prediction; Ajay Agrawal, Joshua S. Gans, and Avi Goldfarb; https://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.33.2.31
● Definition of Robotic Process Automation, UiPath, https://www.uipath.com/rpa/robotic-process-automation
● https://www.expert.ai/blog/machine-learning-definition/
● https://www.geeksforgeeks.org/machine-learning/
● https://www.predictionmachines.ai/
Self reflection
Watch the below documentary about AI and its impact on human life.
Reflect on the impact and benefits of AI on society.
Exercises
Exercise: Multiple choice
and answer the question below.
