L
Layla El Asri
Researcher at Microsoft
Publications - 38
Citations - 1454
Layla El Asri is an academic researcher from Microsoft. The author has contributed to research in topics: Reinforcement learning & Context (language use). The author has an hindex of 18, co-authored 36 publications receiving 1137 citations. Previous affiliations of Layla El Asri include Georgia Institute of Technology & Georgia Tech Lorraine.
Papers
More filters
Proceedings ArticleDOI
Frames: A Corpus for Adding Memory to Goal-Oriented Dialogue Systems
Layla El Asri,Hannes Schulz,Shikhar Sharma,Jeremie Zumer,Justin Harris,Emery Fine,Rahul Mehrotra,Kaheer Suleman +7 more
TL;DR: A rule-based baseline is proposed and the frame tracking task is proposed, which consists of keeping track of different semantic frames throughout each dialogue, and the task is analysed through this baseline.
Posted Content
Relevance of Unsupervised Metrics in Task-Oriented Dialogue for Evaluating Natural Language Generation
TL;DR: An empirical study indicates that automated metrics such as BLEU have stronger correlation with human judgments in the task-oriented setting compared to what has been observed in the non task- oriented setting.
Posted Content
TextWorld: A Learning Environment for Text-based Games
Marc-Alexandre Côté,Ákos Kádár,Xingdi Yuan,Ben Kybartas,Tavian Barnes,Emery Fine,Jim A. Moore,Ruo Yu Tao,Matthew Hausknecht,Layla El Asri,Mahmoud Adada,Wendy Tay,Adam Trischler +12 more
TL;DR: TextWorld as mentioned in this paper is a Python library that handles interactive play-through of text games, as well as backend functions like state tracking and reward assignment, allowing users to handcraft or automatically generate new games.
Book ChapterDOI
TextWorld: A Learning Environment for Text-Based Games
Marc-Alexandre Côté,Ákos Kádár,Xingdi Yuan,Ben Kybartas,Tavian Barnes,Emery Fine,Jim A. Moore,Matthew Hausknecht,Layla El Asri,Mahmoud Adada,Wendy Tay,Adam Trischler +11 more
TL;DR: TextWorld is a Python library that handles interactive play-through of text games, as well as backend functions like state tracking and reward assignment, and comes with a curated list of games whose features and challenges the authors have analyzed.
Posted Content
Policy Networks with Two-Stage Training for Dialogue Systems
TL;DR: This paper shows that, on summary state and action spaces, deep Reinforcement Learning (RL) outperforms Gaussian Processes methods and shows that a deep RL method based on an actor-critic architecture can exploit a small amount of data very efficiently.