Do you really know about AI? — Part 2

Daniel Deutsch
Feb 9 · 5 min read
Photo by Helena Lopes —

More and more people are getting into the field of machine learning and AI. There are many ways to acquire knowledge in this field. But do you really know about the basics in the field of AI? During my studies, I came across various topics and formulated questions. Can you answer them? This is part 2 of a series of questions.

5.Machine Learning

  1. What is learning?
  2. What is Machine Learning?
  3. Elaborate Mitchell’s definition of Machine Learning (1997)?
  4. Name one practical example of the use of Machine Learning.
  5. How is a learning element defined?
  6. Name three types of feedback in Machine Learning.
  7. What is inductive learning? What is its difficulty in terms of Machine Learning?
  8. What is Ockham’s Razor in regard of the inductive learning method?
  9. What findings produced the famous pigeons in Skinner box experiment?
  10. What are neural networks?
  11. How were they discovered?
  12. Name the components of an artificial neuron.
  13. What is a perceptron?
  14. What is the use of the activation function in a perceptron? What is the relation to Boolean functions?
  15. Explain the learning method of a perceptron.
  16. How are weights optimized in a perceptron?
  17. Part of the learning process is an error. How can it be measured and how can it be minimized?
  18. What are the differences between a Threshold Activation Function and a Sigmoid Activation Function? Why is the later more popular?
  19. What are Multilayer Perceptrons?
  20. Describe the information processing of them.
  21. What is backpropagation learning?
  22. What are the training steps when minimizing a network error?
  23. What is overfitting? Elaborate on Training- /Test Set Error.
  24. What is Deep Learning? What are the key requirements?
  25. What is a Convolutional Neural Network? What is a convolution?
  26. Why is it so successful in image processing?
  27. What are Recurrent Neural Networks (RNN)?
  28. What is the use of Long-Short Term Memory (LSTM) in the context of RNN?
  29. What is end-to-end learning in this context?
  30. For what type of problems do RNN provide good solutions?
  31. Name three real-life use cases.

6.Machine Learning in Games

  1. What is reinforcement learning? What is the algorithmic approach?
  2. Explain the concept of reinforcement learning in menace by referencing the tic-tac-toe game.
  3. What is the credit assignment problem? What is its solution?
  4. Why are games of chance closer to real-life?
  5. Why are minimax or alpha-beta algorithms not suitable for those types of games?
  6. What is the exploration / exploitation trade-off?
  7. What is a solution to this problem and what are chance nodes?
  8. What issues arise when using chance nodes at a deeper node level?
  9. What is a possible solution of exact probabilities for the outcomes of chance nodes are not known?
  10. What is the Monte-Carlo Sampling?
  11. What is a simulation search? What is the algorithmic approach?
  12. What is the Monte-Carlo search? What is the algorithmic approach?
  13. What is the policy for maximizing your gain in a multi-armed bandit problem?
  14. What is the Upper Confidence Bound Algorithm (UCB)?
  15. What is the Monte-Carlo Tree Search (MCTS)? Explain the algorithmic approach.
  16. What two policies are used in the MCTS?
  17. Describe the process of selective search in the MCTS.
  18. What is UCT Search? What is the difference between MCTS in its algorithmic approach?
  19. In what game did the UCT Search cause a breakthrough? What were the key components and learning steps?
  20. What is the difference between AlphaGo and AlphaZero?
  21. What is the use of Opponent Modeling?

7.Knowledge and Reasoning

  1. What are the four main components of IBM Watson’s architecture?
  2. Who laid the foundation for formal logic?
  3. What is Propositional Logic?
  4. What is First-Order Logic? What is the difference between Propositional Logic?
  5. What is Prolog and Datalog?
  6. What three main components make up Logical Notation?
  7. Why is logic not enough for knowledge representation?
  8. What is Forward Chaining? Elaborate on the Elementary Production Principle (EPP) in this context.
  9. What is Backward Chaining? What is its advantage over Forward Chaining?
  10. What is an ontology?
  11. What is a semantic network?
  12. What is Feigenbaum’s Knowledge Acquisition Bottleneck?
  13. What is the difference between deductive and inductive reasoning?
  14. What does OWL stand for in the context of reasoning?
  15. Name four strategies for knowledge base construction.
  16. Name one domain-specific expert system.
  17. Name one manual knowledge base system. Why is manual KB construction difficult?
  18. Name one collaborative knowledge base system.
  19. Name one automated knowledge base system.
  20. What is the Semantic Web? What is the difference to the www?
  21. What is linked open data in this context?


  1. What is planning?
  2. What is a representation in the context of planning?
  3. Why is it difficult to find suitable heuristics for a state in a planning problem?
  4. How do you use First-Order Logic in this case and what is the main issue with this approach?
  5. What is situation calculus?
  6. What is the frame problem in situation calculus and how do action rules relate to it?
  7. What is STRIPS?
  8. What are STRIPS operators?
  9. Illustrate an example using STRIPS semantics.
  10. What is progression planning?
  11. What is regression planning?
  12. Why is regression planning often better than progression planning?
  13. What is the inverse action application?
  14. Good heuristics are key for a state-space search. What are two approaches to find a good search heuristic?
  15. What is the Sussman Anomaly?
  16. What is Partial Order Planning? Name an example.
  17. What is the difference between state-space planning and plan-space planning?
  18. What domain-independent heuristics can be used in each case?
  19. What are causal links and ordering constraints?
  20. How can conflict between those recognized and resolved?


Daniel is an entrepreneur, software developer, and lawyer. His knowledge and interests evolve around business law and programming machine learning applications. To the core, he considers himself a problem solver of complex environments, which is reflected in his various projects. Don’t hesitate to get in touch if you have ideas, projects or problems.

Connect on:

Createdd Notes

Articles on Programming and Law

Daniel Deutsch

Written by

Business Law and Machine Learning. Pushing the limits to make the world a better place. Open for Projects of any kind.

Createdd Notes

Articles on Programming and Law

More From Medium

More on Machine Learning from Createdd Notes

More on Machine Learning from Createdd Notes

Do you really know about AI? — Part 3

More on Knowledge from Createdd Notes

More on Knowledge from Createdd Notes

Do you really know about AI? — Part 1

More from Daniel Deutsch

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade