What are tutorials in conferences?

Question

In addition to the main sessions, the workshops, and the special sessions, many conferences provide also some tutorials, usually the day before the beginning of the event or the day after the end of the event.

Computer science NIPS conference, for example, is one of these.

I've attended some of them, and I know what they are. But, if I talked with colleagues about tutorials, each of them would give a different definition of this conference aspects.

What actually are tutorials in conferences?

Are they specific lectures for young attendees of the conference?

Are they specific talks to give the guests the opportunity to explain and advertize a particular aspect of their research?

Or are they just a way to add one day to the conference schedule?

Answer 1

I think this varies a lot from field to field. In High-performance Computing and Scientific Computing, tutorials are very likely to be a set of lectures and hands-on laboratory sessions giving attendees instruction in how to use a computer, API, or other software. This could include "Programming for Intel Xeon Phi", "Introduction to Computing System X", "Python for HPC", etc. (many of which I have personally been involved in teaching). These kinds of tutorials are typical of conferences like Supercomputing and IEEE Cluster.

Answer 2

At the conferences I attend, the tutorials tend to be sort of an oral survey of a topic, introducing practitioners (young or old) to a coherent body of work likely to be of interest to them. For example, a person might give a detailed "how to" introduction to a new (or newly imported) set of methods that many people are interested in applying, or might present a coherent picture of important results within a subfield.