Ph.D, TSENDSUREN Munkhdalai - Empasoft Institute of Technology

TSENDSUREN Munkhdalai

Ph.D

Machine Learning & Artificial Intelligence

Natural Language Processing & Speech Processing

Degree Obtained

Fields of Research

Profile

Education:

Aug 2017: University of Massachusetts, USA (Postdoc)
Aug 2015: Graduated from the Chungbuk National University, South Korea (PhD degree)
Aug 2015: Graduated from the University of Montreal, Canada (Deep Learning Summer School)
Aug 2012: Graduated from the Chungbuk National University, South Korea (Master’s degree)
Jun 2010: Graduated from the National University of Mongolia (Bachelor degree)

Experience:

Oct 2020 - Present: Senior Research Scientist at Google, USA
Sep 2017 - Oct 2020: Senior Researcher at Microsoft Research Montreal, Canada
Oct 2015 - Aug 2017: Postdoctoral Research at University of Massachusetts, USA
Dec 2012 - Feb 2014: Research Engineer at Waagle Startup, USA (remote and part time)
Feb 2010 - Jun 2010: Software engineer at Green Systems LLC, Mongolia (part time)
Jun 2009 - Feb 2010: Software engineer at xAjax LLc, Mongolia (part time)

Selected publications

Diverse Distributions of Self-Supervised Tasks for Meta-Learning in NLP. EMNLP, 2021
Learning Associative Inference Using Fast Weight Memory. ICLR, 2021
Self-supervised meta-learning for few-shot natural language classification tasks. Conference on Empirical Methods in Natural Language Processing. (EMNLP), 2020
Exploring and predicting transferability across nlp tasks. Conference on Empirical Methods in Natural Language Processing. (EMNLP), 2020
Metalearned Neural Memory. Conference on Neural Information Processing Systems (NeurIPS), 2019
Building Dynamic Knowledge Graphs from Text using Machine Reading Comprehension. ICLR, 2019
Rapid adaptation with conditionally shifted neurons. ICML, 2018
Meta Networks. ICML, 2017
Reasoning with Memory Augmented Neural Networks for Language Comprehension. ICLR, 2017
Neural Semantic Encoders. EACL, 2017
Neural Tree Indexers for Text Understanding. EACL, 2017
A deep neural network architecture for class imbalance problem in binary classification. Knowledge-Based Systems. 2020
Clinical relation extraction toward drug safety surveillance using electronic health record narratives: classical learning versus deep learning. JMIR public health and surveillance 4 (2). 2018
Incorporating domain knowledge in chemical and biomedical named entity recognition with word representations. Journal of Cheminformatics, 2015
Self-training significance space of support vectors for imbalanced biomedical event data. BMC Bioinformatics, 2015

Short Bio

I am a research scientist at Google. Before Google, I was a researcher at Microsoft Research Montreal. I got my MSc and PhD under the excellent supervision of Prof. Keun Ho Ryu from the Department of Computer Science at Chungbuk National University, South Korea. I studied my bachelor in Ulaanbaatar, Mongolia at the National University of Mongolia where my thesis topic was on demand forecasting with regression analysis. My current research focus is on meta-learning, memory and attention systems, rapid and temporal adaptations and learning feedbacks in artificial neural networks. As a graduate student, I had a chance to work for a couple of startups where we focused on personalized document processing pipeline and search engine at scale. From 2012-2015, I actively participated in Kaggle competitions and I peaked at world rank 422 (top <1.0%).

CONTACTS

LOCATION