Contrary to popular belief, the new paradigm based on data-driven KPIs is less about technology and more about science. You can pretty much buy any technology, but your ability to adapt to an even more digital future depends on developing the next generation of skills, closing the gap between talented-AI supply and demand.
Our ability to successfully work is strictly related to the development of our artificial intelligence skills, a humanistic momentum filled by science supporting humans in doing, not doing, or even preferring that thing rather than another. This momentum can be considered as an agent in tune with the next generation of transversal and pervasive competencies, exactly as electricity or water do: while the first one ignites what surrounds us, the second can reach the thinnest and deepest rifts.
If you’ve ever taken a TV remote control when you were a baby, your brain at an abstract level has been using reinforcement learning to learn the best potential buttons combination circumstances in a wide array of states that maximize your ability to watch a beautiful children’s cartoon. Reinforcement learning (that’s without the “deep”) involves an agent exploring an environment many times over, gradually learning which actions it should take in different situations in order to achieve its goals.
Its reliance on autonomous exploration and learning by trial and error make it very different from other forms of machine learning. In recent years, deep-learning techniques using neural networks have been used in conjunction with reinforcement learning, resulting in a novel approach called “deep-reinforcement learning.” Neural networks can better model high-level abstractions during the learning process and combining the two techniques together have yielded state-of-the-art results across many problem areas. The RL algorithm’s policy is another important concept, representing how an algorithm decides which action to take in a given state. The policy involves selecting an action that helps maximize rewards over the long term. The agent is able to improve its policy over time by correlating which actions lead to achieving greater feedbacks progressively.
Deep-reinforcement learning is not different from this framework but instead uses neural networks to model the policy during training. These networks are also increasingly being used to model the state in environments such as art auctions, dating apps, on-demand platforms, and other daily life contexts, where unlike remote control, that has a limited set of buttons for a finite set of channels, the machine has to process people’s emotions to understand what will happen in any given point in time. Our emotions, indeed, are the result of what we have become, from year 0 until this moment. They are the complex of behaviors that makes us different from anyone else but similar to someone; a sort of storage of our behavioral paths, that we have been used to go through since even before having memory. In the same way, capturing passions allows us to switch what is valid for one person to many, therefore to understand more audiences and to permeate the whole, just as the analogy of water we mentioned above. While up until now AR was applied to physical things (e.g. showing the inside limbs of the human body during a surgical operation), we are now dealing with a broad concept that slides into the most intangible, fascinating world of emotions. The AI looks at images not only understanding and analyzing physical aspects, but also being able to catch the non-physical sides, meaning the feelings that hide behind that image or the thrill that an image will give birth to in the observer.
In a nutshell – it is a matter of going beyond pure rationality to capture the passion: understanding the key to finding the passion for beauty is much more of a fluid and enhanced ability than that of knowing how to capture the rationality of men. While rationality lets you fall into bias and mental traps, it makes you “lose the game”, following your heart can make you score over and over.