14 Conclusions

 

This chapter covers

  • Important takeaways from this book
  • The limitations of deep learning
  • Possible future directions for deep learning, machine learning, and AI
  • Resources for further learning and applying your skills in practice

You’ve almost reached the end of this book. This last chapter will summarize and review core concepts while also expanding your horizons beyond what you’ve learned so far. Becoming an effective AI practitioner is a journey, and finishing this book is merely your first step on it. I want to make sure you realize this and are properly equipped to take the next steps of this journey on your own.

We’ll start with a bird’s-eye view of what you should take away from this book. This should refresh your memory regarding some of the concepts you’ve learned. Next, I’ll present an overview of some key limitations of deep learning. To use a tool appropriately, you should not only understand what it can do but also be aware of what it can’t do. Finally, I’ll offer some speculative thoughts about the future evolution of deep learning, machine learning, and AI. This should be especially interesting to you if you’d like to get into fundamental research. The chapter ends with a short list of resources and strategies for further learning about machine learning and staying up to date with new advances.

14.1 Key concepts in review

14.1.1 Various approaches to AI

14.1.2 What makes deep learning special within the field of machine learning

14.1.3 How to think about deep learning

14.1.4 Key enabling technologies

14.1.5 The universal machine learning workflow

14.1.6 Key network architectures

14.1.7 The space of possibilities

14.2 The limitations of deep learning

14.2.1 The risk of anthropomorphizing machine learning models

14.2.2 Automatons vs. intelligent agents

14.2.3 Local generalization vs. extreme generalization

14.2.4 The purpose of intelligence

Final words