TEXT AND TEXT AND AUTOMATED BIASES AUTOMATED BIASES NATURAL - PowerPoint PPT Presentation

TEXT AND TEXT AND AUTOMATED BIASES AUTOMATED BIASES

NATURAL LANGUAGES ARE THE NATURAL LANGUAGES ARE THE BASE HUMAN COMMUNICATION BASE HUMAN COMMUNICATION we learn from books of all kinds about complex topics and keep ourself updated

USEFULL APPLICATIONS USEFULL APPLICATIONS structure big amounts of text (by topics or certain words) understand the meaning of text voice recognition text generation (summaries, q&a systems)

COMMON TASKS COMMON TASKS AllenNLP demos Spacy demos

HOW DO WE MAKE COMPUTERS HOW DO WE MAKE COMPUTERS TRY TO UNDERSTAND TRY TO UNDERSTAND LANGUAGE? LANGUAGE? The langauge of each person is different Language is ambigious Language requires contextual information it's constantly evolving

APPROACHES IN THE PAST APPROACHES IN THE PAST 1. Rule based 2. probabilistic models and linear classifiers. 3. deep learning

DEEP LEARNING DEEP LEARNING

HOW TO DEAL WITH SEQUENCES HOW TO DEAL WITH SEQUENCES

DEEP LEARNING FOR NLP DEEP LEARNING FOR NLP from symbolic representations to tensors/vectors and embeddings how to represent words in the input layer?

ONE HOT ENCODING ONE HOT ENCODING word : n dimensions (for dictionary size) car : 1 0 0 ... 0 dog : 0 1 0 ... 0 cat : 0 0 1 ... 0 apple : 0 0 0 ... 1 scales bad no relationship between words no context/semantic information

WORD EMBEDDINGS WORD EMBEDDINGS much less dimensions then words in the dictionary relationship between words build from training language models

TRAINING WORD VECTORS TRAINING WORD VECTORS (WORD EMBEDDINGS) (WORD EMBEDDINGS)

LANGUAGE MODELS LANGUAGE MODELS Predicting the next character / word in a sequence

The Unreasonable Effectiveness of Recurrent Neural Networks

WORD ASSOCIATIONS WORD ASSOCIATIONS

Demo time

SUMMARY SUMMARY Language models builds Word Embeddings Word Embeddings are Word representations in tense spaces The contain semantical information about a word Association are reflected in the relationships of words

SUMMARY SUMMARY Language models builds Word Embeddings Word Embeddings are Word representations in tense spaces The contain semantical information about a word Association are reflected in the relationships of words There are problematic associations

Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings

Word Embedding Association Test Semantics derived automatically from language corpora necessarily contain human biases

Biases are consolidated Historical bias Are Emily and Greg More Employable than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination underepresented group

QUESTIONS TO ASK QUESTIONS TO ASK Who build the model From what dataset was it build Where is the model used?

REAL WORLD APPLICATIONS AND REAL WORLD APPLICATIONS AND THERE PROBLEMS THERE PROBLEMS GOOGLE TRANSLATE GOOGLE TRANSLATE

Google Translate Keeps Spitting Out Creepy Religious Prophecies

CHATBOTS CHATBOTS

Can virtual humans be more engaging than real ones?

MICROSOFTS CHATBOT TAY MICROSOFTS CHATBOT TAY

WOEBOT WOEBOT

HOW EXTREME BIAS BECOMES WHEN HOW EXTREME BIAS BECOMES WHEN FED WITH BAD DATA FED WITH BAD DATA

Norman A.I

Bias is identical to meaning, and it is impossible to employ language meaningfully without incorporating human bias.

THANK YOU THANK YOU Get in touch transfluxus@posteo.de twitter.com/ramin__