# Perspectives

This is a bit more bloviating, but it is a write-up of a discussion I had with the students once at the start and end of the course, and I think it is particularly useful.

I, like a lot of other folks, tend to enjoy being able to sort things into groups as I am learning new things. The sort of mental "aha" moment you have when you realize "oh this is like this other thing I know" often in association with "but it's slightly different". [Genus-differentia](https://en.wikipedia.org/wiki/Genus%E2%80%93differentia_definition) is a common way of defining things in general. 

It turns out that when learning a broad topic, like deep learning, it is actually incredibly useful to build these mental maps of related things, understanding how they are related, but also the subtle differences. 

### **Supervised, Unsupervised, Semi-supervised learning**

Of course, the fundamental way of dividing up what we are learning is based on the kind of data we have and how we learn from them. This 

### Inductive bias

There's a bit of personal bias (no pun intended) here, as much of my research has been on applying (or not applying) inductive bias on scientific problems. Inductive bias is the set of assumptions (like symmetry) you provide to your model to steer it towards a particular solution. The interplay of inductive bias based on the 

### Modality

This goes hand in hand with inductive bias (to the extent it could even be considered a sub-category), but learning on the different modalities, like text, image, audio, video, and graphs, informs how we develop our model and learning techniques. 


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://szaman.gitbook.io/intro-to-deep-learning/readme/perspectives.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.