EXAM PREPARATION

Natural Language Interaction

* What is a conversation turn? Explain the turn taking rules.
* Explain (and illustrate) what collaboration in dialogue means. What are collaborative
  responses?
* List and explain the Gricean maxims of cooperation.
* What are adjacency pairs in dialogue, what characteristics do they have, and how is the
  concept of adjacency pair useful in dialogue modelling?
* What is the general global structure of a conversation? Give some typical examples of the 
  respective parts.
* Explain the notion of a speech act.
* Explain the notion of common ground. When is a piece of information considered common
  ground between dialogue participants?
* What is the process of grounding?
* What levels of interpretation does Clark (1996) distinguish? Explain what they mean and
  illustrate with examples.
* What are the various reasons for grounding problems (why can an agent have difficulties
  grounding an utterance)? Explain and illustrate.
* What (three) kinds of verification strategies can a grounding model of a dialogue system
  implement? Discuss advantages and disadvantages.
* Explain the difference between finite state machine-based, frame-based and informationstate
  update based dialogue models. What are their advantages and disadvantages? Give
  examples of applications in which you would use them.
* Explain the concepts behind information state update based dialogue modelling. What
  kinds of rules have to be defined in such a model?
* What is the difference between fixed and mixed initiative? What advantages and
  disadvantaged has fixed system initiative in comparison to mixed initiative model?
* Explain the concepts of technical evaluation, usability evaluation, and customer evaluation.
  Give some (at least 3) questions that users may be asked to assess the usability of a
  system. 
* What are basic motivations to carry out WoZ experiments? What are the challenges in
  conducting a WoZ experiment? What requirements must a WoZ experiment meet in order
  to serve the purpose of a simulated human-machine interaction? Give examples for
  methods or tools that are employed to meet the requirements.
* Reinforcement Learning (RL) is based on the concept of Markov Decision Process (MDP).
  Give a short and informal description of the components of an MDP and the relevant
  notions (i.a. the Markov Property and the three kinds of rewards). Explain how MDP is used
  in RL in general. RL for natural languages interaction uses recorded dialogue data for
  training. There are two kinds of information derived from the empirical data which is fed to
  the learning process. What are they?
* Explain the difference between model-based and simulation-based reinforcement learning.
* Explain the concepts of exploration vs. exploitation in reinforcement learning.

* Dialogue evaluation: Explain the concepts of technical evaluation,
usability evaluation, and customer evaluation. Give some (at least 3)
questions that users may be asked to assess the usability of a system.

* WoZ experiments: What are the challenges in conducting a WoZ
experiment? Give examples for methods that help to meet the basic
requirements on a WoZ system.

* Reinforcement learning: What are the basic kinds of information that
go into the computation of the expected cumulative reward (according
to Bellman?s equation)? You need not reproduce the equation literally,
but can give an informal description.

* MDP: Given an ISU-based dialogue manager, like in the MP3-Player
example of the lecture slides, but for train information, and a
dialogue state where the system has understood that the user wants to
go to Hamburg. If th next system ation/utterance is:

S: Which day do you want to go to Hamburg?

What could happen next, and how would this be modeled in the MDP
framework? Give an informal description in terms of state change and
reward.

* Q-Learning: What is the motivation for the step-size parameter, and
what is the standard way to handle it?

* Input interpretation: Input interpretation for dialogue can be done
by recognition grammars which are enriched with interpretation tags,
and by interpretation grammars.  Explain the basic difference.