Events

EMNLP w/ Juha, Maciej & Julien

LOUHI 2020: The 11th International Workshop on Health Text Mining and Information Analysis.

20 November | 09:30 | Register here →

Simple Hierarchical Multi-Task Neural End-To-End Entity Linking for Biomedical Text

Recognising and linking entities is a crucial first step to many tasks in biomedical text analysis, such as relation extraction and target identification. Traditionally, biomedical entity linking methods rely heavily on heuristic rules and predefined, often domain-specific features. The features try to capture the properties of entities and complex multi-step architectures to detect, and subsequently link entity mentions. We propose a significant simplification to the biomedical entity linking setup that does not rely on any heuristic methods. The system performs all the steps of the entity linking task jointly in either single or two stages. We explore the use of hierarchical multi-task learning, using mention recognition and entity typing tasks as auxiliary tasks. We show that hierarchical multi-task models consistently outperform single-task models when trained tasks are homogeneous. We evaluate the performance of our models on the biomedical entity linking benchmarks using MedMentions and BC5CDR datasets. We achieve state-of-theart results on the challenging MedMentions dataset, and comparable results on BC5CDR.


Maciej Wiatrak
Machine Learning Engineer

Maciej is a Machine Learning Engineer at BenevolentAI, where he focuses on Natural Language Processing (NLP) and its applications in drug discovery. In the past he has worked in research labs at University College London and University of Edinburgh on topics such as multi-agent reinforcement learning and generative adversarial networks. Recently, he has become interested in geometric deep learning and utilising different manifolds for effective representation learning.

Juha Iso-Sipila
Senior Lead AI Scientist

Juha has MSc (1996) and PhD (2008) degrees in Computer Science from Tampere University of Technology. After graduation he worked for Nokia in Finland and in China, developing speech recognition, text-to-speech synthesis and predictive text input technologies for Nokia. In 2013 Juha relocated to London to work as Head of Research at SwiftKey. During this time SwiftKey launched a neural network based mobile predictive text input system. Juha has been at BenevolentAI from 2017. His main interests are in Deep Learning for biomedical NLP and distributed computing using Apache Spark.


SustaiNLP 2020

20 November | Time TBC | Register here →

Learning Informative Representations of Biomedical Relations with Latent Variable Models

Extracting biomedical relations from large cor-pora of scientific documents is a challengingnatural language processing task.  Existing ap-proaches  usually  focus  on  identifying  a  rela-tion either in a single sentence (mention-level)or across an entire corpus (pair-level). In bothcases, recent methods have achieved strong re-sults by learning a point estimate to representthe relation;  this is then used as the input toa relation classifier.  However, the relation ex-pressed  in  text  between  a  pair of  biomedicalentities is often more complex than can be cap-tured by a point estimate.  To address this is-sue, we propose a latent variable model withan arbitrarily flexible distribution to representthe relation between an entity pair.  Addition-ally, our model provides a unified architecturefor both mention-level and pair-level relationextraction.    We  demonstrate  that  our  modelachieves results competitive with strong base-lines for both tasks while having fewer param-eters and being significantly faster to train. Wemake our code publicly available.

You can find the paper published here, and the code can be accessed on GitHub here.

The lead author of this paper was Harshil Shah, our Machine Learning Intern.

Julien Fauqueur
Lead AI Scientist

Julien Fauqueur is a Lead AI Scientist at BenevolentAI. He has developed natural language processing methods for biomedical information extraction and, since this year, he has been the AI Lead for the Chemistry group. He holds a PhD in computer science from INRIA in France. While a research associate at the University of Cambridge, then as a senior researcher in industry, he developed new methods and products using computer vision and machine learning. He co-authored around 30 academic publications and 5 patents.

More Posts

You Might Also Like

Blog
Intern at BenevolentAI part I: meet our 2020 intern cohort
What impactful work did our interns get up to across Engineering, Data Science, ML and business operations this summer? Get to know them and their work in our tech internships blog.
Nov 26, 2020
News
FDA grants Emergency Use Authorisation for baricitinib in hospitalised COVID-19 patients nine months after initial hypothesis was published by BenevolentAI
BenevolentAI scientists first identified baricitinib as a potential treatment for COVID-19 in early February 2020 using Benevolent's AI tools and biomedical knowledge graph.
Nov 20, 2020
News
BenevolentAI at NeurIPS 2020: Machine Learning in Drug Discovery
BenevolentAI is happy to announce it is sponsoring NeurIPS 2020. Join us to hear about data diversity and ML applied drug discovery, and to learn about careers in the field.
Nov 17, 2020
Blog
Careers with Impact: 5 learnings from machine learning applied drug discovery
Last week, we brought together four of our exceptional colleagues for a panel discussion on careers in machine learning applied drug discovery. Here are some of our main takeaways:
Nov 17, 2020
Blog
Data published in Science Advances shows baricitinib reduces COVID-19 morbidity and mortality
Research published in Science Advances supports BenevolentAI’s AI-generated hypothesis from late January for baricitinib as a treatment for COVID-19.
Nov 13, 2020
News
Sir Nigel Shadbolt joins BenevolentAI as a non-executive director
BenevolentAI strengthens its Board with the appointment of AI pioneer Sir Nigel Shadbolt as Non-Executive Director.
Nov 3, 2020