Search Results for author:
Found 5 papers, 2 papers with code
Sparse-IFT: Sparse Iso-FLOP Transformations for Maximizing Training Efficiency
Recent research has focused on weight sparsity in neural network training to reduce FLOPs, aiming for improved efficiency (test accuracy w. r. t training FLOPs).
SPDF: Sparse Pre-training and Dense Fine-tuning for Large Language Models
In this work, we show the benefits of using unstructured weight sparsity to train only a subset of weights during pre-training (Sparse Pre-training) and then recover the representational capacity by allowing the zeroed weights to learn (Dense Fine-tuning).
RevBiFPN: The Fully Reversible Bidirectional Feature Pyramid Network
However, training them requires substantial accelerator memory for saving large, multi-resolution activations.
Ranked #310 on
Image Classification
on ImageNet
(using extra training data)
Approximating Wisdom of Crowds using K-RBMs
We propose a method to aggregate noisy labels collected from a crowd of workers or annotators.
DAiSEE: Towards User Engagement Recognition in the Wild
We introduce DAiSEE, the first multi-label video classification dataset comprising of 9068 video snippets captured from 112 users for recognizing the user affective states of boredom, confusion, engagement, and frustration in the wild.
