Talk
Curriculum Learning: Scores, Plans, Dynamics, and NLP
Abstract: Curriculum learning is an effective and natural strategy in human learning. It plays an important role in challenging tasks such as language learning. However, current machine learning (ML) paradigms are mostly built upon repeatedly practicing the same training data/tasks with a random order, which is non-adaptive to the learning process. Moreover, they do not plan multiple learning stages in advance as humans.
Lunch will be served starting at 12:15. (Vegan options availalble, but GF options could be a bit odd... like a lettuce wrap. Let us know if you have other dietary restrictions!)
Sources of Bias in Psycholinguistic Data
The 12th annual Language Science Day is scheduled for Friday, October 28, 2022.
Language Science Day (LSD) is a signature annual event for Maryland's language science community, bringing together up to 200 students and faculty from across the university and affiliated centers. Participants get to know their fellow language scientists, exchange ideas, showcase their research, and discover opportunities for training or collaboration.
Following Instructions and Asking Questions
Abstract: As we move towards the creation of embodied agents that understand natural language, several new challenges and complexities arise for grounding (e.g. complex state spaces), planning (e.g. long horizons), and social interaction (e.g. asking for help or clarifications). In this talk, I'll discuss improvements to embodied instruction following within ALFRED and initial steps towards building agents that ask questions or model theory-of-mind.
Meg Cychosz (HESP) will give an overview of the process of applying for academic jobs (applications, interviews, job talks, negotiations), and answer all your burning questions!
Lunch will be served starting at 12:15. (Vegan options availalble, but GF options could be a bit odd... like a lettuce wrap.)
Enhanced Perception and Cognition in Deaf Sign Language Users? EEG and Behavioral Evidence
Planet Cloze: The expanding universe of lexical prediction models
Rosa Lee (LING), Katherine Howitt (LING), London Dixon (LING), Tal Ness (HESP), Masato Nakamura (LING)
Zoom link: https://umd.zoom.us/j/9057393501
Join College of Education faculty for a virtual roundtable discussion on ADVANCING Linguistic Justice.
Panelists:
- Ana Taboada Barber (CHSE)
- Shenika Hankerson (TLPL)
- Laura Mahalingappa (TLPL)
- Min Wang (HDQM)
Moderator: Jen Turner (College of Education ADVANCE Professor)
Pages
