Artificial Intelligence in 2020 and Beyond
Author | Min Sun, Chief AI Scientist, Appier
2020 has a great ring to it. In lots of ways, it has for a long time been a signifier of ‘the future’. Many researchers and experts have, over the past decades, projected 2020 as a landmark year when we might expect any multitude of things to happen or come to a head in some way.
And now ‘the future’ is here. We don’t yet have flying cars – or even self-driving cars in any widely-adopted way – but we have seen technology advance in leaps and bounds in other ways in recent decades. As computing power continues to grow, we can expect to see this momentum increase.
One area that continues to see rapid advancement is artificial intelligence (AI). There have been some critical breakthroughs in AI technology within the last decade that have allowed AI to be applied in truly revolutionary ways in both business and society – take medical diagnostics for example – and I’d like to share some thoughts on how we might expect to see AI continue to transform the way we do things in 2020.
AI Will Be More Strategic and Able to Act on Its Own
To date, AI has become extremely good at analyzing data to make predictions. Human decision-makers then take these outcomes and determine how to proceed. However, in the coming year and beyond, we can expect to see that AI will not only predict what happens next, but also act on that information itself. Some of these decisions will require extremely efficient use of data, and AI’s continued ability to learn from just a small amount of data.
Let’s look at an example in digital marketing. By looking at past behavior of customers, it’s possible to predict whether someone might make an online purchase. That said, the behavioral data available for someone who makes purchases frequently will be different (and more abundant) from the data available for someone who makes purchases infrequently, or perhaps never, on your website.
AI will get better at identifying these differences and acting based on what it learns. Moreover, AI can go beyond the relatively simple task of predicting what frequent buyers will do and start to explore the unique pattern of hesitant or infrequent buyers on its own, suggesting actions to take to engage them.
AI Will Be Able to Consider Multiple Objectives
One major difference between artificial and human intelligence is that currently, most AI programs are very good at optimizing one single objective (i.e. achieving one goal), whereas humans are typically able to balance multiple objectives and modify that balance on the fly as they move towards a goal.
For instance, driving from location A to location B is one obvious single goal, but other goals such as how fast, how safe, and how comfortable the drive should be are also implicitly considered by human drivers. We can expect to see AI become better at balancing multiple goals at once.
Traditionally, AI is trained with one quantifiable goal or a goal that manually combines multiple goals or objectives. Newer AI can be trained to optimize multiple objectives and learn to explore the trade-off between objectives. Such trained AI can be controlled by human users to operate at a specific trade-off.
For instance, a user can control an AI ‘driver’ to go to A from B in under 30 minutes with maximum comfort. Another user can control the AI driver to go to A from B in under 10 minutes while sacrificing comfort as little as possible. This is an advance toward human-centered AI, which makes AI easier for people to control.
As noted, humans are typically able to manage multiple priorities and outcomes and weigh them against each other and now, AI will be able do this at greater speed and scale. In general, we can expect to see all AI systems designed with the human end-user in mind, allowing for optimal collaboration between artificial and human intelligence.
Advances in Distributed Learning
Data is king, and optimal collection, management, accessibility and unification is going to continue to be critical to be able to effectively leverage it in the real world. As the amount of data grows and the types of data become more disparate – text, video, imagery, audio, etc. – it is going to be vital for industries, organizations and governments to develop effective data ecosystems that share information.
This means that distributed learning – the ability for AI systems to solve problems by learning from a broad set of existing and shared data that doesn’t need to be relocated or reacquired – will become increasingly important. We can certainly expect to see this when it comes to, for example, initiatives like smart cities where policy makers will want to leverage data without moving it around to better understand things like transportation patterns, road safety, and so on.
Overall, the ability for AI to do more is good news for business and society. Humans will be able to work more efficiently, with tools at their fingertips that allow them to do their jobs faster and more accurately. This also gives people time to focus on the things that are still best suited for humans, such as creative development and empathetic and thoughtful communications.
AI also encourages boundless innovation in the ways it can be applied, and as it becomes capable of doing more, AI practitioners and researchers around the world will no doubt be able to find exciting and useful ways to use it, improving how we live, work and play.
* This article was originally published on CDO Trends.
WE ARE HERE TO HELP
YOU MIGHT ALSO LIKE
2019 is the tenth year since Alibaba took the reins of Singles’ Day – an event originating in China in the 1990s to celebrate people who are not in romantic relationships – and turned it into a major online shopping event. Any entity that sells online highly anticipates Singles’ Day and events like it, such as Cyber Monday which traditionally follows the American Thanksgiving holiday. They prepare to break through the ‘noise’ and attract as many shoppers as possible with a goal of increased sales. Last year, we here at Appier shared some advice for brands and retailers about how they can use artificial intelligence (AI) to prepare for the run-up to Singles’ Day to pique the interest of consumers before the day itself. This year, we’d like to help marketers answer the following question using machine learning: “How do brands and retailers convert the mass media exposure and high cost of Singles’ Day into improved customer lifetime value?” For big campaigns like Singles’ Day, overwhelmed marketers are typically only able to focus on high-level execution that drives reach and single campaign gross merchandise volume (GMV). There is typically no bandwidth or resource during these events to look at longer-term
Broadly speaking, there are two types of machine learning algorithms: supervised and unsupervised learning. Supervised learning is the more common of the two, and is typically easier to implement than unsupervised learning. What Is Supervised Learning? Supervised learning algorithms are designed to learn by example. They are used when the human practitioner knows the answer to a problem, and wants to train the AI to be able to find it out. It is like learning with the assistance of a teacher, guiding the algorithm towards the ‘correct’ answer, as opposed to an unsupervised learning algorithm, which is like a child learning on their own by experimentation and trial and error. To train a supervised learning algorithm, you will need to pair a set of inputs with specific outputs. The algorithm will then search for patterns within the inputs to correlate with the outputs. Based on this training data, the supervised learning algorithm can then take in unseen inputs and determine which label to assign them. The aim? To predict the correct label for newly presented input data in order to categorize it and make sense of it. Supervised learning: · Is simpler and more common than unsupervised learning ·
We are glad to have the opportunity to speak to renowned AI experts who are putting their exceptional knowledge into practice to solve some of society’s most pressing issues. In our past couple of conversations, we have spoken quite a bit about the role of AI in transforming business, but what about its role in helping to solve major societal issues like natural disaster prevention and preparation? In Part 3 of our ongoing video series, Appier’s Chief AI Scientist Dr. Min Sun sat down with Professor Naonori Ueda, Deputy Director, RIKEN Center for Advanced Intelligence Project. He is also the Director of Machine Learning, Data Science Center, NTT Communication Science Laboratories and a visiting professor for the Graduate School of Informatics at both Kyoto and Kobe Universities. Professor Ueda has been working in the fields of artificial intelligence and machine learning for more than 30 years, and is passionate about applying his research to real-world issues that affect everyone. In this discussion, the two AI professors talk about the importance of artificial intelligence experts working with domain experts to ensure that every discipline can make use of AI research to solve problems in almost every area imaginable. Professor Ueda uses