https://machinelearningmastery.com/sequence-prediction/

Predictions with Sequences

Sequence prediction is used to predict the next value of a given input sequence. One example is product recommendations, which predicts a customer's next purchase given a sequence of their past purchases. Another example is stock market prediction.

Sequence Prediction Models

Sequence classification predicts a class label for a given input sequence. Sentiment analysis is one example, which takes a sequence of text and predicts the label for the sentiment (positive or negative). 

Sequence Classification Models

Recurrent neural networks, or RNNs, are a family of neural networks for processing sequential data. A recurrent neural network is specialized for processing a sequence of inputs $x^{(1)}, . . . , x^{(τ)}$ and each time add additional layers of comprehension on top of the previous inputs. 


Recurrent Neural Network (RNN)

You have trained word embeddings using a text dataset of $$m_1$$ words. You are considering using these word embeddings for a language task, for which you have a separate labeled dataset of $$m_2$$ words. Keeping in mind that using word embeddings is a form of transfer learning, under which of these circumstance would you expect the word embeddings to be helpful?

Sequence Model Question #1

In the word2vec algorithm, you estimate $$P\left(t|c\right)$$, where t is the target word and c is a context word. How are t and c chosen from the training set?

Sequence Model Question #2

Suppose you learn a word embedding for a vocabulary of 10000 words. Then the embedding vectors should be 10000 dimensional, so as to capture the full range of variation and meaning in those words.

Sequence Moel Question #4

In beam search, if you increase the beam width B, which of the following would you expect to be true? Check all that apply.

Sequence Model Question #3

Tokenization is the process of converting a sequence of text into smaller units, known as tokens. It is a foundational step in Natural Language Processing, and there are numerous different methods and strategies for how a text can be tokenized.

Tokenization

In sequence modeling architectures, standard notation is used to represent the inputs and outputs. A source sequence is typically denoted as $$x_1...x_m$$, where $$m$$ represents the length of the source sequence. Its corresponding sequence of embedding vectors is denoted as $$\mathbf{e}_{1}^{x}...\mathbf{e}_{m}^{x}$$. Similarly, the target sequence that the model aims to generate is denoted as $$y_1...y_n$$, with length $$n$$, and its corresponding embedding sequence is represented as $$\mathbf{e}_{1}^{y}...\mathbf{e}_{n}^{y}$$.

Notation for Source and Target Sequences

In sequence modeling, interpolation and extrapolation represent two fundamentally different challenges with a significant gap in difficulty. Interpolation involves estimating values that fall within the temporal range of data the model has already observed, while extrapolation requires predicting future, unseen values beyond the observed range. Because extrapolation demands the model generalize beyond its training distribution, it is substantially harder than interpolation. This asymmetry has a critical practical implication: when working with sequential data, one must always respect the temporal order of observations during training—that is, a model should never be trained on future data to predict the past.

Interpolation vs. Extrapolation in Sequence Models

For causal sequence models—those in which time progresses naturally forward—estimating values in the forward direction (predicting the future from past observations) is typically much easier and more practical than estimating in the reverse direction (reconstructing the past from future observations). This asymmetry arises because causal models are designed to capture how past events generate future outcomes, making the forward conditional distribution the natural object of estimation. Attempting to invert this process is inherently more challenging because the mapping from effects back to causes is often one-to-many and ill-conditioned.

Forward vs. Reverse Direction Estimation

The initial step in preparing text for sequence modeling is to read the raw text from a dataset into memory as a continuous string. For example, a complete text file, such as a book, can be downloaded and loaded as a single sequence of characters before any further processing occurs.

Reading Raw Text for Sequence Data

Sequence models are statistical supervised learning functions designed to process, predict, or classify based on sequence data. A fundamental task in this domain is to estimate the joint probability of an entire sequence. When dealing with natural language data composed of discrete tokens, such as words, these estimated functions are commonly referred to as language models. Sequence models provide the capacity to evaluate the likelihood of sequences (for example, comparing the naturalness of outputs from translation systems), sample new sequences, and optimize for the most likely sequence outputs.

University of Michigan - Ann Arbor

Google

Claude

Supervised learning learns a function mapping inputs x to outputs y using labeled training examples (x,y). Linear regression, logistic regression, and neural networks are examples of supervised learning algorithms.

Supervised Learning

Reference of Foundations of Large Language Models Course

Dive into Deep Learning

Which of the following are use-cases of supervised learning?

Methods of supervised statistical learning

- Classification Problems
- Regression Problems

Types of supervised learning problems

Use cases of supervised statistical learning

Which ones are true about Supervised statistical learning?

Which of the following are examples of supervised machine learning? Select all that apply.

Categories of supervised learning algorithms

A Basic Supervised Statistical Learning Workflow

- Training set: to train the model
- Validation set (development set): to tune the model parameters
- Test set: to test the performance of the model on unobserved data

Division of dataset in supervised statistical learning

Feature scaling greatly affects which of the following supervised machine learning methods? 

1. Supervised learning allows you to collect data or produce a data output from the previous experience
2. Helps you to optimize performance criteria using experience
3 Supervised machine learning helps you to solve various types of real-world computation problems.

Adavantages of supervised learning

1. Decision boundary might be overtrained if your training set which doesn't have examples that you want to have in a class
2. You need to select lots of good examples from each class while you are training the classifier.
3. Classifying big data can be a real challenge.
4.Training for supervised learning needs a lot of computation time.


Disadvantages of Supervised Learning

1. Before doing anything else, you need to decide what kind of data is to be used as a training set
2. You need to decide the structure of the learned function and learning algorithm.
3. Gathere corresponding outputs either from human experts or from measurements

Best practices for Supervised Learning


The Supervised Learning Workshop: A New, Interactive Approach to Understanding Supervised Learning Algorithms, 2nd Edition Paperback – February 28, 2020
https://www.amazon.com/Supervised-Learning-Workshop-Interactive-Understanding/dp/1800209045/ref=sr_1_1_sspa?dchild=1&keywords=Supervised+Learning&qid=1613836679&sr=8-1-spons&psc=1&spLa=ZW5jcnlwdGVkUXVhbGlmaWVyPUExQzMzWkdMTDlBVldRJmVuY3J5cHRlZElkPUEwMTY1NTYxMUNKMk8zN1YwR0ZZTyZlbmNyeXB0ZWRBZElkPUEwMjIwMDEwMjRUUzk1MllXUU5DSSZ3aWRnZXROYW1lPXNwX2F0ZiZhY3Rpb249Y2xpY2tSZWRpcmVjdCZkb05vdExvZ0NsaWNrPXRydWU=

The Supervised Learning Workshop: A New, Interactive Approach to Understanding Supervised Learning Algorithms

Sequence Models

The purpose of a statistical learning model is to use a statistical model and the training data to find the function $\hat{f}$ as an estimate of the unknown function $f$ that maps the matrix of input variables $X$ to the outcome vector $Y$, as best as possible.
$$Y = f(X) + \epsilon \approx \hat{f}(X)$$ 

Purpose of supervised statistical learning

In supervised learning, input values represent the set of known features or attributes provided to a model to make predictions. A critical first step in data preprocessing is partitioning the dataset to separate these input columns from the designated target values that the model aims to estimate.

Input Values

In information retrieval, search ranking is a task that involves imposing a rank on a set of items. Instead of simply determining if an item is relevant to a user's query, the objective is to assign a score to each relevant candidate and display the top-rated elements most prominently, utilizing machine learning and behavioral models to obtain query-dependent relevance scores.

Search Ranking

Sequence learning encompasses machine learning problems that require a model to ingest sequences of inputs, emit sequences of outputs, or both. Unlike standard tasks that process a fixed number of inputs independently, sequence learning models maintain context across successive observations, making them essential for handling varying-length data such as video snippets, text, or patient histories over time.

Sequence Learning

In supervised learning, a target value (often called a label) represents the designated output or objective variable that a model is trained to predict. During data preprocessing, the dataset is typically partitioned to isolate the target column from the input values, enabling the model to learn the mapping from features to the desired prediction.

Target Values

In standard supervised learning, we commonly rely on the independent and identically distributed (IID) assumption. This principle dictates that both the training and test data are drawn independently from the exact same underlying probability distribution, denoted as $$P(X,Y)$$. Without this strong assumption that $$P(\cdot) = Q(\cdot)$$ (where $$Q$$ is the test distribution), it would be impossible to justify using patterns learned from the training data to make predictions on new test data.

Independent and Identically Distributed (IID) Assumption

In traditional supervised learning applications, the output y was usually an integer or a real number.

Traditional Supervised Learning Outputs

Linear regression, logistic regression, and neural networks are examples of supervised learning algorithms.

Learn Before

Related

Learn After