Deep Learning and Explainable AI

1. Deep Learning and Explainable AI

Previously, you learned about many types of machine learning, such as supervised machine learning, clustering, time series analysis, and NLP. Finally, we're ready to discuss two of the hottest topics in machine learning: Deep Learning and Explainable AI.

2. What is Deep Learning?

So what is Deep Learning? Deep Learning, also sometimes called "neural networks" or "neural nets", is a special type of machine learning that can solve more complex problems. It requires much, much more data than traditional machine learning. It is best used in cases where the inputs are less structured, such as large amounts of text or images.

3. Explainable AI

One of the main drawbacks of Deep Learning is a lack of explainability. Although Deep Learning can make very accurate predictions, it's not always clear why the model is making a specific prediction. Methods that allow us to understand the factors that lead to each prediction are also known as "Explainable AI", where "AI" stands for "Artificial Intelligence".

4. Case Study: Explainable AI

Let's examine a typical problem in Explainable AI. Suppose we are investigating customer cancellations using a traditional machine learning model. Our trained model can tell us two things. First, it can predict whether or not a given customer is likely to churn. Second, it can tell us which features were important in making this decision. This is the "explainable" part. This additional explainability can provide important insights. For instance, we might learn that certain demographics are much more likely to cancel their subscriptions. Our Marketing and Customer Support teams can now use this insight to change their outreach strategies and address this deficiency.

5. Case Study: Inexplicable AI

Contrast that example with a typical Deep Learning problem. Suppose we want to recognize hand-written letters. We don't really care why a particular image was classified as an "a", as long as the predictions are highly accurate. Deep Learning is a perfect solution to this problem because we don't care about explainability and we probably have a large, image-based set of training data.

6. When to use Deep Learning

Let's generalize what we've learned. Before we choose Deep Learning as a solution, we should ask ourselves a few questions. One, does our training data have many features or is it difficult to understand it as a simple array of features? Data such as images or text are particularly suitable to Deep Learning. Two, do we have a very large amount of training data? Deep Learning requires more training data than traditional machine learning. If you don't have millions of examples, Deep Learning might not work for your company. And, finally, does the model need to be predictive or explanatory? Deep Learning is great for predictive modeling, but can leave us perplexed if we care about why each prediction was made. Simpler models might have less predictive power but can be better when clarity is essential.

7. Let's practice!

Now that we understand the difference between Deep Learning and Explainable AI, let's practice!

Create Your Free Account

By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.