Constrained Mean Shift Using Distant Yet Related Neighbors for Representation Learning

1 code implementation • 8 Dec 2021 • KL Navaneet, Soroush Abbasi Koohpayegani, Ajinkya Tejankar, Kossar Pourahmadi, Akshayvarun Subramanya, Hamed Pirsiavash

On the other hand, far away NNs may not be semantically related to the query.

Pseudo Label Representation Learning +1

A Simple Baseline for Low-Budget Active Learning

1 code implementation • 22 Oct 2021 • Kossar Pourahmadi, Parsa Nooralinejad, Hamed Pirsiavash

However, most such methods assume that a large subset of the data can be annotated.

Active Learning Image Classification +1

TaxoNN: A Light-Weight Accelerator for Deep Neural Network Training

no code implementations • 11 Oct 2020 • Reza Hojabr, Kamyar Givaki, Kossar Pourahmadi, Parsa Nooralinejad, Ahmad Khonsari, Dara Rahmati, M. Hassan Najafi

In this work, first we present a novel approach to add the training ability to a baseline DNN accelerator (inference only) by splitting the SGD algorithm into simple computational elements.