Date of Submission

Spring 2016

Academic Programs and Concentrations

Computer Science

Project Advisor 1

Sven Anderson

Abstract/Artist's Statement

Deep learning is a relatively new area in the field of machine learning, and its full potential has yet to be known. Nevertheless, systems employing deep learning have made significant strides in the computing world - beating professionals in Go, a game previously thought to be too difficult for current computers to solve, accurately describing images and identifying a speaker by voice. The goal of this project is to evaluate how well recurrent neural networks, a variant of deep learning, perform against humans in simplifying single words in sentences. Multiple language models were generated from the recurrent neural network, and the best models were used to rank lexical choices. Predictions made by the network were compared with human-generated lexical rankings.

Access Agreement

On-Campus only

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

This work is protected by a Creative Commons license. Any use not permitted under that license is prohibited.

Bard Off-campus Download

Bard College faculty, staff, and students can login from off-campus by clicking on the Off-campus Download button and entering their Bard username and password.