Machine Learning for Finance电子书

Machine Learning for Finance

Machine Learning for Finance

Why subscribe?

Packt.com

Contributors

About the author

About the reviewer

Preface

Who this book is for

What this book covers

To get the most out of this book

Download the example code files

Download the color images

Conventions used

Get in touch

Reviews

1. Neural Networks and Gradient-Based Optimization

Our journey in this book

What is machine learning?

Supervised learning

Unsupervised learning

Reinforcement learning

The unreasonable effectiveness of data

All models are wrong

Setting up your workspace

Using Kaggle kernels

Running notebooks locally

Installing TensorFlow

Installing Keras

Using data locally

Using the AWS deep learning AMI

Approximating functions

A forward pass

A logistic regressor

Python version of our logistic regressor

Optimizing model parameters

Measuring model loss

Gradient descent

Backpropagation

Parameter updates

Putting it all together

A deeper network

A brief introduction to Keras

Importing Keras

A two-layer model in Keras

Stacking layers

Compiling the model

Training the model

Keras and TensorFlow

Tensors and the computational graph

Exercises

Summary

2. Applying Machine Learning to Structured Data

The data

Heuristic, feature-based, and E2E models

The machine learning software stack

The heuristic approach

Making predictions using the heuristic model

The F1 score

Evaluating with a confusion matrix

The feature engineering approach

A feature from intuition – fraudsters don't sleep

Expert insight – transfer, then cash out

Statistical quirks – errors in balances

Preparing the data for the Keras library

One-hot encoding

Entity embeddings

Tokenizing categories

Creating input models

Training the model

Creating predictive models with Keras

Extracting the target

Creating a test set

Creating a validation set

Oversampling the training data

Building the model

Creating a simple baseline

Building more complex models

A brief primer on tree-based methods

A simple decision tree

A random forest

XGBoost

E2E modeling

Exercises

Summary

3. Utilizing Computer Vision

Convolutional Neural Networks

Filters on MNIST

Adding a second filter

Filters on color images

The building blocks of ConvNets in Keras

Conv2D

Kernel size

Stride size

Padding

Input shape

Simplified Conv2D notation

ReLU activation

MaxPooling2D

Flatten

Dense

Training MNIST

The model

Loading the data

Compiling and training

More bells and whistles for our neural network

Momentum

The Adam optimizer

Regularization

L2 regularization

L1 regularization

Regularization in Keras

Dropout

Batchnorm

Working with big image datasets

Working with pretrained models

Modifying VGG-16

Random image augmentation

Augmentation with ImageDataGenerator

The modularity tradeoff

Computer vision beyond classification

Facial recognition

Bounding box prediction

Exercises

Summary

4. Understanding Time Series

Visualization and preparation in pandas

Aggregate global feature statistics

Examining the sample time series

Different kinds of stationarity

Why stationarity matters

Making a time series stationary

When to ignore stationarity issues

Fast Fourier transformations

Autocorrelation

Establishing a training and testing regime

A note on backtesting

Median forecasting

ARIMA

Kalman filters

Forecasting with neural networks

Data preparation

Weekdays

Conv1D

Dilated and causal convolution

Simple RNN

LSTM

The carry

Recurrent dropout

Bayesian deep learning

Exercises

Summary

5. Parsing Textual Data with Natural Language Processing

An introductory guide to spaCy

Named entity recognition

Fine-tuning the NER

Part-of-speech (POS) tagging

Rule-based matching

Adding custom functions to matchers

Adding the matcher to the pipeline

Combining rule-based and learning-based systems

Regular expressions

Using Python's regex module

Regex in pandas

When to use regexes and when not to

A text classification task

Preparing the data

Sanitizing characters

Lemmatization

Preparing the target

Preparing the training and test sets

Bag-of-words

TF-IDF

Topic modeling

Word embeddings

Preprocessing for training with word vectors

Loading pretrained word vectors

Time series models with word vectors

Document similarity with word embeddings

A quick tour of the Keras functional API

Attention

Seq2seq models

Seq2seq architecture overview

The data

Encoding characters

Creating inference models

Making translations

Exercises

Summary

6. Using Generative Models

Understanding autoencoders

Autoencoder for MNIST

Autoencoder for credit cards

Visualizing latent spaces with t-SNE

Variational autoencoders

MNIST example

Using the Lambda layer

Kullback–Leibler divergence

Creating a custom loss

Using a VAE to generate data

VAEs for an end-to-end fraud detection system

VAEs for time series

GANs

A MNIST GAN

Understanding GAN latent vectors

GAN training tricks

Using less data – active learning

Using labeling budgets efficiently

Leveraging machines for human labeling

Pseudo labeling for unlabeled data

Using generative models

SGANs for fraud detection

Exercises

Summary

7. Reinforcement Learning for Financial Markets

Catch – a quick guide to reinforcement learning

Q-learning turns RL into supervised learning

Defining the Q-learning model

Training to play Catch

Markov processes and the bellman equation – A more formal introduction to RL

The Bellman equation in economics

Advantage actor-critic models

Learning to balance

Learning to trade

Evolutionary strategies and genetic algorithms

Practical tips for RL engineering

Designing good reward functions

Careful, manual reward shaping

Inverse reinforcement learning

Learning from human preferences

Robust RL

Frontiers of RL

Multi-agent RL

Learning how to learn

Understanding the brain through RL

Exercises

Summary

8. Privacy, Debugging, and Launching Your Products

Debugging data

How to find out whether your data is up to the task

What to do if you don't have enough data

Unit testing data

Keeping data private and complying with regulations

Preparing the data for training

Understanding which inputs led to which predictions

Debugging your model

Hyperparameter search with Hyperas

Efficient learning rate search

Learning rate scheduling

Monitoring training with TensorBoard

Exploding and vanishing gradients

Deployment

Launching fast

Understanding and monitoring metrics

Understanding where your data comes from

Performance tips

Using the right hardware for your problem

Making use of distributed training with TF estimators

Using optimized layers such as CuDNNLSTM

Optimizing your pipeline

Speeding up your code with Cython

Caching frequent requests

Exercises

Summary

9. Fighting Bias

Sources of unfairness in machine learning

Legal perspectives

Observational fairness

Training to be fair

Causal learning

Obtaining causal models

Instrument variables

Non-linear causal models

Interpreting models to ensure fairness

Unfairness as complex system failure

Complex systems are intrinsically hazardous systems

Catastrophes are caused by multiple failures

Complex systems run in degraded mode

Human operators both cause and prevent accidents

Accident-free operation requires experience with failure

A checklist for developing fair models

What is the goal of the model developers?

Is the data biased?

Are errors biased?

How is feedback incorporated?

Can the model be interpreted?

What happens to models after deployment?

Exercises

Summary

10. Bayesian Inference and Probabilistic Programming

An intuitive guide to Bayesian inference

Flat prior

<50% prior

Prior and posterior

Markov Chain Monte Carlo

Metropolis-Hastings MCMC

From probabilistic programming to deep probabilistic programming

Summary

Farewell

Further reading

General data analysis

Sound science in machine learning

General machine learning

General deep learning

Reinforcement learning

Bayesian machine learning

Other Books You May Enjoy

Leave a review - let other readers know what you think

Index