Caglar Gulcehre
Caglar Gulcehre researches machine learning for language, reasoning and sequential decisions. His work includes gated recurrent networks and memory-based language learning. Gulcehre co-authored a study with Junyoung Chung, Kyunghyun Cho and Yoshua Bengio comparing gated recurrent units, LSTM units and traditional recurrent units. The experiments covered polyphonic music and speech-signal modelling. He earned a MILA doctorate supervised by Yoshua Bengio. He defended Learning and time: on using memory and curricula for language understanding in 2018. His research includes reinforcement learning, offline reinforcement learning, large deep-learning architectures and representation learning. His CLAIRE group develops efficient and reliable reasoning algorithms.
Timeline
-
2014
Evaluating gated recurrent networks
Gulcehre co-authored a study with Junyoung Chung, Kyunghyun Cho and Yoshua Bengio comparing gated recurrent units, LSTM units and traditional recurrent units. The experiments covered polyphonic music and speech-signal modelling.
-
2018
Memory and language learning
He earned a MILA doctorate supervised by Yoshua Bengio. He defended Learning and time: on using memory and curricula for language understanding in 2018.
-
As of October 2026
Reinforcement learning and representations
His research includes reinforcement learning, offline reinforcement learning, large deep-learning architectures and representation learning. His CLAIRE group develops efficient and reliable reasoning algorithms.
Sources and credits
Photo: ellis.eu; Caglar Gulcehre profile photograph; creator not stated. License: Not stated by source. Source: ellis.eu. Cropped to a square and resized. Ask us to remove or replace this photo.