Open AccessPosted Content
Fraternal Dropout
TL;DR: A simple technique called fraternal dropout is proposed that takes advantage of dropout to train two identical copies of an RNN (that share parameters) with different dropout masks while minimizing the difference between their (pre-softmax) predictions.
read more
Abstract: Recurrent neural networks (RNNs) are important class of architectures among neural networks useful for language modeling and sequential prediction. However, optimizing RNNs is known to be harder compared to feed-forward neural networks. A number of techniques have been proposed in literature to address this problem. In this paper we propose a simple technique called fraternal dropout that takes advantage of dropout to achieve this goal. Specifically, we propose to train two identical copies of an RNN (that share parameters) with different dropout masks while minimizing the difference between their (pre-softmax) predictions. In this way our regularization encourages the representations of RNNs to be invariant to dropout mask, thus being robust. We show that our regularization term is upper bounded by the expectation-linear dropout objective which has been shown to address the gap due to the difference between the train and inference phases of dropout. We evaluate our model and achieve state-of-the-art results in sequence modeling tasks on two benchmark datasets - Penn Treebank and Wikitext-2. We also show that our approach leads to performance improvement by a significant margin in image captioning (Microsoft COCO) and semi-supervised (CIFAR-10) tasks.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Do Not Have Enough Data? Deep Learning to the Rescue!
Ateret Anaby-Tavor,Boaz Carmeli,Esther Goldbraich,Amir Kantor,George Kour,Segev Shlomov,Naama Tepper,Naama Zwerdling +7 more
- 03 Apr 2020
TL;DR: This work uses a powerful pre-trained neural network model to artificially synthesize new labeled data for supervised learning and shows that LAMBADA improves classifiers' performance on a variety of datasets.
Recent Advances in Deep Learning: An Overview
Matiur Rahman Minar,Jibon Naher +1 more
TL;DR: This paper is going to briefly discuss about recent advances in Deep Learning for past few years.
144
IAUnet: Global Context-Aware Feature Learning for Person Reidentification
TL;DR: A novel block, interaction-aggregation-update (IAU), for high-performance person reID that is lightweight, end-to-end trainable, and can be easily plugged into existing CNNs to form IAUnet, which performs favorably against state of the art on both image and video reID tasks.
57
Improving visual question answering using dropout and enhanced question encoder
TL;DR: This paper proposes coherent dropout and siamese dropout mechanism to solve the two problems of co-adaptations of neurons and the explosion of output variance in VQA, and develops an enhanced question encoder called Multi-path Stacked Residual RNNs which is deeper and wider and more powerful than current shallow question encoding.
37
Towards Effective and Generalizable Fine-tuning for Pre-trained Molecular Graph Models
TL;DR: Two straightforward yet effective strategies to attain better generalization performance are proposed: MolAug, which enriches the molecular datasets of down-stream tasks with chemical homologies and enantiomers; WordReg, which controls the complexity of the pre-trained models with a smoothness-inducing regularization built on dropout.
27
References
•Posted Content
Deep Residual Learning for Image Recognition
TL;DR: This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.
117.9K
Long short-term memory
TL;DR: A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
99K
•Journal Article
Dropout: a simple way to prevent neural networks from overfitting
TL;DR: It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.
•Posted Content
Empirical evaluation of gated recurrent neural networks on sequence modeling
TL;DR: These advanced recurrent units that implement a gating mechanism, such as a long short-term memory (LSTM) unit and a recently proposed gated recurrent unit (GRU), are found to be comparable to LSTM.
14.1K
•Proceedings Article
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
Kelvin Xu,Jimmy Ba,Ryan Kiros,Kyunghyun Cho,Aaron Courville,Ruslan Salakhudinov,Ruslan Salakhudinov,Rich Zemel,Rich Zemel,Yoshua Bengio,Yoshua Bengio +10 more
- 06 Jul 2015
TL;DR: An attention based model that automatically learns to describe the content of images is introduced that can be trained in a deterministic manner using standard backpropagation techniques and stochastically by maximizing a variational lower bound.