I have graduated from CMU! I am now a DPhil candidate in computer science at the University of Oxford and research scientist at Google DeepMind .

I was a second-year research master's student at the Language Technologies Institute , School of Computer Science , Carnegie Mellon Univeristy . I am very fortunate to be advised by Chris Dyer, and also collaborate closely with Noah Smith , Miguel Ballesteros, and Graham Neubig. I am a Graduate Research Assistant working on the DARPA Low Resource Languages for Emergent Incidents ( LORELEI ) project here at CMU.

In the summer of 2016 I did a research internship at the University of Washington , Seattle, advised by Noah Smith . Before coming to CMU I was a research intern at the Nara Institute of Science and Technology Computational Linguistics lab, Japan, advised by Kevin Duh. I completed an M.Sc. in Computer Science from the University of Oxford in 2014, advised by Phil Blunsom, and a B.Eng. in Informatics from Institut Teknologi Bandung in 2013.

Please see my CV (last updated December 2016) for more details.

News

• [10/2017] Invited talk at the University of Cambridge Computer Laboratory
• [09/2017] Joined the University of Oxford Computer Science Department and DeepMind
• [08/2017] Invited talk at the University of Pennsylvania CLunch Talk
• [04/2017] Won an Outstanding Long Paper award at EACL 2017

Contact

adhiguna.kuncoro [ at ] cs [ dot ] ox [ dot ] ac [ dot ] uk

Research Interests

I am working on the intersection between machine learning and natural language processing (NLP).

I am broadly interested in the following:

• Dependency parsing, phrase-structure parsing, named entity recognition (and other structured prediction problems)
• Neural network and representation learning methods for NLP tasks
• Cross-lingual and multi-lingual learning for low-resource languages
• Generative models of language and semi-supervised learning with latent variables
• Reinforcement learning for structured prediction

Publications

Conference Proceedings

• Memory Architectures In Recurrent Neural Network Language Models .
  Dani Yogatama, Yishu Miao, Gabor Melis, Wang Ling , Adhiguna Kuncoro , Chris Dyer , Phil Blunsom .
  In Proc. ICLR 2018, Vancouver, Canada
  


• What Do Recurrent Neural Network Grammars Learn About Syntax? .
  Adhiguna Kuncoro, Miguel Ballesteros, Lingpeng Kong, Chris Dyer, Graham Neubig , Noah A. Smith.
  In Proc. EACL 2017 (Long Paper), Valencia, Spain
  Outstanding Paper Award


• Distilling an Ensemble of Greedy Dependency Parsers into One MST Parser.
  Adhiguna Kuncoro, Miguel Ballesteros, Lingpeng Kong, Chris Dyer, Noah A. Smith.
  In Proc. EMNLP 2016 (Long Paper), Austin, TX, USA
  Code


• Recurrent Neural Network Grammars.
  Chris Dyer, Adhiguna Kuncoro, Miguel Ballesteros, Noah A. Smith.
  In Proc. NAACL 2016 (Long Paper), San Diego, CA, USA
  Code

Workshop Proceedings

• Cross Modal Content Based Objective for Learning Adequate Multimodal Representations.
  Adhiguna Kuncoro ^* , Akash Bharadwaj^*, Seungwhan Moon^*, Volkan Cirik ^*, Chris Dyer, Louis-Philippe Morency. In Proc. ICML 2016 Workshop on Multi-View Representation Learning (MVRL), Oral Presentation, New York, NY, USA
  ^* indicates equal contributions.

Preprints

• Dependency Parsing with LSTMs: An Empirical Evaluation.
  Adhiguna Kuncoro, Yuichiro Sawai, Kevin Duh, Yuji Matsumoto. ArXiv

Personal

My first name ('Adhi') is pronounced as 'Uh-dee'. I grew up in Jakarta, Indonesia, before moving to Bandung, West Java for my undergraduate years. I have additionally lived in Melbourne, Oxford, Nara, and Pittsburgh. I enjoy Scuba diving.

If you are an Indonesian student looking to apply/enroll at CMU, feel free to send me an e-mail and check out our webpage. We'd be happy to help! I'd also recommend the Indonesia Mengglobal webpage and their mentorship program.