More than 100 people joined us for our 4th MLT Women in Machine Learning, this time it was held at the brand new Google Japan office in the impressive Shibuya Stream building. We heard two fantastic talks – one on production-level ML, the other from a research perspective – and had a great panel discussion afterwards with interesting questions from the audience. A fun networking session with drinks and finger food concluded the afternoon at Google.
OO Design Patterns for Machine Learning
Object-oriented programming has long been known for its design patterns. In this talk, Chris Gerpheide described a number of design patterns that are particularly useful for machine learning applications, including why and how you can implement them yourself. Find her presentation slides and code snippets here.
Chris Gerpheide is the CTO at Bespoke Inc. Before joining Bespoke, she was engineering manager at Amazon Web Services. In her free time, she enjoys learning Japanese, hiking, and teaching children programming.
Detecting word meaning change – from probabilistic models to BERT
Word usage, meaning and connotation change throughout time, mirroring the cultural and technological evolution of society. Diachronic word embeddings are used to grasp these changes in an unsupervised way. It is useful for social and linguistic research, by detecting and interpreting the causes of semantic shifts, but also for many NLP tasks to study temporal corpora. Syrielle Montariol introduced methods to detect word meaning change, from models relying on classical word embeddings to the revolution of contextualised embeddings. Find Syrielle’s slides here: Detecting Word Meaning Change (PDF).
Syrielle Montariol received a Master’s degree in Statistics from a French engineering school and started her PhD in 2018 in Paris, working in parallel at Université Paris-Saclay and at the company Société Générale. Her research goal is to use semantic change as a tool to detect and understand social and political crisis.
We would like to thank Google Japan for sponsoring this event. If your company is interested to sponsor the next MLT Women in Machine Learning event, please reach out to email@example.com.
Support our Open Education efforts and help us keep MLT events for free for participants by becoming a Patron: https://www.patreon.com/MLTOKYO
Join the community and find code, resources and workshops here: