Learning objectives

A motivating example

[Withheld from all public copies]

Anomymized data

• Generally we anonymize data because, while the data itself is broadly useful, providing full information could harm others or oneself
• Examples:
  • Studying criminal behavior use can create a list of people with potentially uncaught criminal offenses
    • If one retains a list of identities, then there is an ethical dilemma:
      • Protect study participants by withholding the list
      • Provide the list to the government
        • This harms future knowledge generation by sowing distrust
    • Solution: Anonymous by design
  • Website or app user behavior data
    • E.g.: FiveThirtyEight’s Uber rides dataset

What could go wrong if the Uber data wasn’t anonymized?

Anonymization is tricky

Both Allman & Paxson, and Partridge warn against relying on the anonymisation of data since deanonymisation techniques are often surprisingly powerful. Robust anonymisation of data is difficult, particularly when it has high dimensionality, as the anonymisation is likely to lead to an unacceptable level of data loss [3]. – TPHCB 2017

• There are natural limits to anonymization, particularly when there is a limited amount of potential participants in the data
  • Example: Web browser tracking at Panopticlick

Responsibilities generating data

• Keep users as unidentifiable as feasible
• If you need to record people’s private information, make sure they know
  • This is called informed consent
• If you are recording sensitive information, consider not keeping identities at all
  • Create a new, unique identifier (if needed)
  • Maintain as little identifying information as necessary
  • Consider using encryption if sensitive data is retained
  • Can unintentionally lead to infringements of human rights if the data is used in unintended ways

Informed consent

• When working with data about people, they should be informed of this and consent to the research, unless the data is publicly available
• From SMU’s IRB Handbook: (2017 SEP 18 version)

Also, note the existence of the PDPA law in Singapore

Human rights

• Recall the drug users example
  • If data was collected without their consent, and if it was not anonymized perfectly, then this could lead to leaking of drug user’s information to others

What risks does this pose? Consider contexts outside Singapore as well.

A novel approach using a GAN

• Source: Learning Anonymized Representations with Adversarial Neural Networks
• On handwriting classification, cases that can be deanonymized drop from 40% to 3.3%
  • Accuracy drops from ~98% down to 95%, a much smaller drop

How the GAN works

• There are 2 general components to a GAN:
  1. A generative network: It’s goal is to generate something from some data
     • In this case, it generates a dataset with good predictability for the “predictor” network AND bad predictability for the “adversarial” network
  2. An adversarial: It’s goal is to thwart the generative network
     • In this case, it tries to determine the writers’ identitities

By iterating repeatedly, the generative network can find a strategy that can generally circumvent the discriminitive network

• Related work in biology: Privacy-preserving generative deep neural networks support clinical data sharing (also GAN based)

Responsibilities using data

“The collection, or use, of a dataset of illicit origin to support research can be advantageous. For example, legitimate access to data may not be possible, or the reuse of data of illicit origin is likely to require fewer resources than collecting data again from scratch. In addition, the sharing and reuse of existing datasets aids reproducibility, an important scientific goal. The disadvantage is that ethical and legal questions may arise as a result of the use of such data” (source)

Responsibilities using data

• Respect for persons
  • Individuals should be treated as autonomous agents
    • People are people
  • Those without autonomy should be protected
• Beneficence
  1. Do not harm (ideally)
  2. Maximize possible benefits and minimize possible harms
     • This can be a natural source of conflict
• Justice
  • Benefits and risks should flow to the same groups – don’t use unwilling or disadvantaged groups who won’t receive any benefit
    • [Extreme] example: Tuskegee Syphilis study

For experiments, see The Belmont Report; for electronic data, see The Menlo Report

R for ML/AI

Python for ML/AI

Others for ML/AI

• C/C++: Also a first class language for TensorFlow!
  • Really fast – precompiled
  • Much more difficult to code in
• Swift: Strong TensorFlow support
• Javascript: Improving support from TensorFlow and others

Why do I keep mentioning TensorFlow?

Why do I keep mentioning TensorFlow?

Other notable frameworks

What are neural networks?

• The phrase neural network is thrown around almost like a buzz word
• Neural networks are actually a specific type class algorithms
  • There are many implementations with different primary uses

What are neural networks?

• Originally, the goal was to construct an algorithm that behaves like a human brain
  • Thus the name
• Current methods don’t quite reflect human brains, however:
  1. We don’t fully understand how our brains work, which makes replication rather difficult
  2. Most neural networks are constructed for specialized tasks (not general tasks)
  3. Some (but not all) neural networks use tools our brain may not have
     • I.e., backpropogation is potentially possible in brains, but it is not pinned down how such a function occurs (if it does occur)

What are neural networks?

• Neural networks are a method by which a computer can learn from observational data
• In practice:
  • They were not computationally worthwhile until the mid 2000s
  • They have been known since the 1950s (perceptrons)
  • They can be used to construct algorithms that, at times, perform better than humans themselves
    • But these algorithms are often quite computationally intense, complex, and difficult to understand
  • Much work has been and is being done to make them more accessible

Types of neural networks

• There are a lot of neural network types
  • See The “Neural Network Zoo”
• Some of the more interesting ones which we will see or have seen:
  • RNN: Recurrent Neural Network
  • LSTM: Long/Short Term Memory
  • CNN: Convolutional Neural Network
  • DAN: Deep Averaging Network
  • GAN: Generative Adversarial Network
• Others worth noting
  • VAE (Variational Autoencoder): Generating new data from datasets

RNN: Recurrent NN

• Recurrent neural networks embed a history of information in the network
  • The previous computation affects the next one
  • Leads to a short term memory
• Used for speech recognition, image captioning, anomaly detection, and many others
  • Also the foundation of LSTM
  • SketchRNN (live demo)

LSTM: Long Short Term Memory

• LSTM improves the long term memory of the network while explicitly modeling a short term memory
• Used wherever RNNs are used, and then some
  • Ex.: Seq2seq (machine translation)

CNN: Convolutional NN

• Networks that excel at object detection (in images)
• Can be applied to other data as well
• Ex.: Inception-v3

DAN: Deep Averaging Network

• DANs are simple networks that simply average their inputs
• Averaged inputs are then processed a few times
• These networks have found a home in NLP
  • Ex.: Universal Sentence Encoder

GAN: Generative Adversarial Network

• Feature two networks working against each other
• Many novel uses
  • Ex.: The anonymization GAN we saw
  • Ex.: Aging images

VAE: Variational Autoencoder

• An autoencoder (AE) is an algorithm that can recreate input data
• Variational means this type of AE can vary other aspects to generate completely new output
  • Good for creating fake data
• Like a simpler, noisier GAN

Motivating examples

What are “vector space models”

• Different ways of converting some abstract information into numeric information
  • Focus on maintaining some of the underlying structure of the abstract information
• Examples (in chronological order):
  • Word vectors:
    • Word2vec
    • GloVe
  • Paragraph/document vectors:
    • Doc2Vec
  • Sentence vectors:
    • Universal Sentence Encoder

Word vectors

• Instead of coding individual words, encode word meaning
• The idea:
  • Our old way (encode words as IDs from 1 to N) doesn’t understand relationships such as:
    • Spatial
    • Categorical
    • Grammatical (weakly when using stemming)
    • Social
    • etc.
  • Word vectors try to encapsulate all of the above
    • They do this by encoding words as a vector of different features

Word vectors: Simple example

• The above is an idealized example
• Notice how we can tell apart different animals based on their relationship with people
• Notice how we can distinguish turkey (the animal) from Turkey (the country) as well

What it retains: word2vec

What it retains: GloVe

How to build word vectors

• Two ways:
  1. Word co-occurrence (like how LDA worked)
     • Global Vectors (GloVe) works this way
       • Available from the text2vec package
  2. Word order (using an NN)
     • word2vec works this way
       • Available from the rword2vec package
       • Uses a 2 layer neural network

How does word order work?

Infer a word’s meaning from the words around it

Refered to as CBOW (continuous bag of words)

How else can word order work?

Infer a word’s meaning by generating words around it

Refered to as the Skip-gram model

An example of using word2vec

• In the BCE paper from Session 6, word2vec was used to provide assurance that the LDA model works reasonably well on annual reports
  1. We trained a word2vec model on random issues of the Wall Street Journal (247.8M words)
  2. The resulting model “understood” words in the context of the WSJ
  3. We then ran a psychology experiment (word intrusion task) on the algorithm

Word intrusion task

• The task is to find which word doesn’t belong
• Each question consisted of 3 words from 1 topic and 1 intruded from another random topic
  • Ex.:
    • Laser, Drug, Viral, Therapeutic
    • Supply, Steel, Capacity, Losses
    • Relief, Lousisiana, Cargo, Assisted

Results

Implementing in R

• A few options:
  • The rword2vec package for word2vec
  • The text2vec package for GloVe
  • Rolling your own neural network for word2vec with keras (guide here)

When are vector embeddings useful?

1. You care about the words used, by not stylistic choices
2. You want to crunch down a bunch of words into a smaller number of dimensions without running any bigger models (like LDA) on the text.

An interactive demo of word similarity

Document vectors

• Document vectors work very similarly to word vectors
  • 1 added twist: a document/paragraph/sentence level factor variable
    • This is used to learn a vector representation of each text chunk
    • Learned simultaneously with the word vectors
      • Caveat: it can also be learned independently using PV-DBOW
• This is quite related to what we learned with LDA as well!
  • Both can tell us the topics discussed

Wikipedia article categorization

Source article (colah.github.io)

Universal Sentence Encoder (USE)

• We saw this briefly last week
  • This is the algorithm with less bias
• Focused on representing sentence-length chunks of text

A fun example of with USE

• Predict Shakespeare with Cloud TPUs and Keras

Cavaet on using USE

• One big caveat: USE only knows what it’s trained on
  • Ex.: Feeding the same USE algorithm WSJ text

Samsung Electronics Co., suffering a handset sales slide, revealed a foldable-screen smartphone that folds like a book and opens up to tablet size. Ah, horror? I play Thee to her alone;
And when we have withdrom him, good all.
Come, go with no less through.

Enter Don Pedres. A flourish and my money. I will tarry. Well, you do!

LADY CAPULET.
Farewell; and you are

How does USE work?

• USE is based on a DAN
  • There is another specification as well
  • Learns the meaning of sentences via words’ meanings
• Learn more: Original paper and TensorFlow site
• In practice, it works quite well

Try it out!

• Run on Google Colab
  • Python code
  • Just click the cells in order, and click run
  • Colab provides free servers to run the code on
    • It still takes a few minutes to run though

Some newer things

• BERT
  • A model structure that’s been winning all sorts of competitions over the past year
    • Based on an architecture called “Transformer”
  • Optimized to mimic question and answer behavior (examples)
  • Now used in Google Search
  • Additional reading
  • Available in TensorFlow Hub
• XLNet
  • Similar objective to BERT, but with a focus on word order

Discussion

What creative uses for the techniques discussed today do you expect to see become reality in accounting in the next 3-5 years?

• Brainstorm with your group and try to come up with 1 good use for some technique discussed today
• Each group will be asked to share 1 use

Recap

Today, we:

• Learned formally what neural networks (NNs) are
• Discussed a variety of NN-based algorithms
• Saw uses for word and sentence vectors in a financial context

For next week

• For next week:
  • Work on the group project!
    • Definitely try to get a submission in on Kaggle
  • We’ll keep talking about neural networks
    • A bit more theory
    • A lot more examples
    • Some real neural networks coded in R

More fun examples

• Interactive:
  • TensorFlow.js examples
  • Semantris
    • A game based on the Universal Sentence Encoder
• Non-interactive
  • Predicting e-sports winners with Machine Learning

Packages used for these slides

• kableExtra
• knitr
• tidyverse
  • dplyr, magrittr, readr

Generating Shakespeare

seed_txt = 'Looks it not like the king?  Verily, we must go! '  # Original code
seed_txt = 'SCENE I. Elsinore. A platform before the Castle.\n\n Enter Francisco and Barnardo, two sentinels.\n\nBARNARDO.\nWho’s there?\n\nFRANCISCO.\nNay, answer me. Stand and unfold yourself.\n\nBARNARDO.\nLong live the King!\n\n'  # Hamlet
seed_txt = 'Samsung Electronics Co., suffering a handset sales slide, revealed a foldable-screen smartphone that folds like a book and opens up to tablet size.'  # WSJ article
# From: https://www.wsj.com/articles/samsung-unveils-foldable-screen-smartphone-1541632221