Chengxi Ye, Chinmaya Devaraj and Michael Maynord win best student paper in IEEE BigData 2018 Workshop
In a recently published paper CS researchers proposed an intuitive method to facilitate constructing artificial neural networks. The method, inspired by human learning, proposes a multi-stage information distillation approach using multi-resolution features and feedbacks. The method facilitates information propagation and makes neural networks more efficient and easier to visualize.
Chengxi Ye, Chinmaya Devaraj, Michael Maynord, Cornelia Fermüller and Yiannis Aloimonos, "Evenly Cascaded Convolutional Networks," 2018 IEEE International Conference on Big Data (Big Data), Seattle, WA, USA, 2018, pp. 4640-4647.
The Department welcomes comments, suggestions and corrections. Send email to editor [at] cs [dot] umd [dot] edu.