Machine Learning Garden

Powered by 🌱Roam Garden

Machine Learning

References

Tags: concept

Sources:

Related notes:

Deep Learning

Reinforcement Learning

Curriculum Learning

Updates:

Notes {{word-count}}

Summary:

Key points:

What is Machine Learning?

In Machine Learning, instead of defining the input -> output relationship by hand, we learn a program that acquires the relationship from data.

Tom Mitchell's definition

A computer program is said to learn from experience EE with respect to some class of tasks TT and performance measure PP, if its performance at tasks in TT, as measured by PP, improves with experience EE.

The Machine Learning method for solving any problem ever.

Define your model class.

Model class refers to the the set of possible programs。

Define your Loss Function.

Loss Function measures if one model in the model class is better than another.

Define your Optimizer.

The optimizer searches the model class to find the model that minimizes the Loss Function.

Run it on a GPU.

Marr's levels of analysis

What is learning?

We want to learn pθ(y∣x)p_\theta (y \mid x), and it is a model which approximates the true p(y∣x)p(y \mid x).

A good model should make the data look probable.

We choose θ\theta such that p(D)=∏ip(xi)pθ(yi∣xi)p(\mathcal{D})=\prod_{i} p\left(x_{i}\right) p_{\theta}\left(y_{i} \mid x_{i}\right) is maximized.

However, one numerical problem here is that we are multiply together many numbers less than one.

To solve the problem, we can use log⁡\log to convert multiplication into addition.

log⁡p(D)=∑ilog⁡p(xi)+log⁡pθ(yi∣xi)=∑ilog⁡pθ(yi∣xi)+ const \log p(\mathcal{D})=\sum_{i} \log p\left(x_{i}\right)+\log p_{\theta}\left(y_{i} \mid x_{i}\right) =\sum_{i} \log p_{\theta}\left(y_{i} \mid x_{i}\right)+\text { const }

θ⋆←arg⁡max⁡θ∑ilog⁡pθ(yi∣xi)\theta^{\star} \leftarrow \arg \max _{\theta} \sum_{i} \log p_{\theta}\left(y_{i} \mid x_{i}\right)

This is called the Maximum Likelihood Estimation (MLE), and it can be formulated as a Negative Log-Likelihood (NLL) problem.

Referenced in

Machine Learning

In Machine Learning, instead of defining the input -> output relationship by hand, we learn a program that acquires the relationship from data.

Deep Learning

Deep Learning is Machine Learning with multiple layers of learned representations.

Machine Learning

The Machine Learning method for solving any problem ever.

Machine Learning