Continuous Learning
CDP, CLV and CX (customer experience) are the first three steps in the process of transforming loyalty using VRM. The fourth step is CL (continuous learning) – creating a feedback loop where new data is ingested by AI-ML systems to create an even better system. Daily, and forever. This is a marketer’s dream – where every action taken by a customer is fed back into the system, and that is used to further personalise the content and experience.
What used to be a slow, time-consuming process by analysts can now be done by machines faster, better and cheaper. This lays the foundation for the 4Rs of customer experience – right message to the right person at the right time on the right channel. CL is at the heart of this world of “omnichannel personalisation”. It is even more important to do this for the most important customers.
So, what exactly is CL? Here is a short overview from Vincenzo Lomonaco:
Continual Learning (CL) is built on the idea of learning continuously and adaptively about the external world and enabling the autonomous incremental development of ever more complex skills and knowledge.
In the context of Machine Learning it means being able to smoothly update the prediction model to take into account different tasks and data distributions but still being able to re-use and retain useful knowledge and skills during time.
The simplest application of CL is in scenarios where the data distributions stay the same but the data keeps coming. This is the classical scenario for an Incremental Learning system.
You can think of a lot of applications like Recommendation or Anomaly Detection systems where data keeps flowing and continually learning from them is really important to refine the prediction model and in the end improve the service offered.
However, nowadays, for most of the commercial DL applications it’s ok to re-train the model from scratch with the cumulated data. The game becomes really interesting instead when the scenario keeps changing over time. This is where Continual Learning really shines and other techniques are unable to solve the problem.
In the world of marketing, the rise of digital customers is creating a continuous stream of data. Machines are best placed to process this and then provide the necessary guidance on the next best actions for every customer. While brands can apply this to all customers, the critical requirement must be to do this for the Best Customers. These 20% customers are where all personalisation efforts need to be super-charged. CL is the way to make systems better daily. If a system becomes 1% better each day, over a year the improvement is 3700%, or 37 times. That’s the power that an AI-ML powered CL system can deliver.
Tomorrow: Part 14
