paper-with-me

홈 › Papers

Breaking the Softmax Bottleneck: A High-Rank RNN Language Model

2017-11-10 · ICLR 2018 1 · Zhilin Yang, Zihang Dai, Ruslan Salakhutdinov, William W. Cohen

We formulate language modeling as a matrix factorization problem, and show that the expressiveness of Softmax-based models (including the majority of neural language models) is limited by a Softmax bottleneck. Given that natural language is highly context-dependent, this further implies that in practice Softmax with distributed word embeddings does not have enough capacity to model natural language. We propose a simple and effective method to address this issue, and improve the state-of-the-art perplexities on Penn Treebank and WikiText-2 to 47.69 and 40.68 respectively. The proposed method also excels on the large-scale 1B Word dataset, outperforming the baseline by over 5.6 points in perplexity.

📄 PDF Abstract BibTeX arXiv:1711.03953

Code (9)

zihangdai/mos 공식 구현 pytorch
cstorm125/thai2fit pytorch
nkcr/overlap-ml pytorch
nunezpaul/MNIST tf
omerlux/NLP-PTB pytorch
omerlux/Recurrent_Neural_Network_-_Part_2 tf
tdmeeste/SparseSeqModels pytorch
yfreedomliTHU/mos-pytorch1.1 pytorch
zhangyaoyuan/GAN-Simplification tf

Tasks

Language ModelingLanguage ModellingVocal Bursts Intensity PredictionWord Embeddings

Methods 이 논문이 사용한 방법론

Sigmoid Activation 설명 없음
Tanh Activation 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Temporal Activation Regularization 설명 없음
Activation Regularization Activation Regularization (AR), or $L\_{2}$ activation regularization, is regularization performed on activations as opposed to weights. It is usually used in conjunction with…
Weight Tying Weight Tying improves the performance of language models by tying (sharing) the weights of the embedding and softmax layers. This…
Embedding Dropout Embedding Dropout is equivalent to performing dropout on the embedding matrix at a word level, where the dropout is broadcast…
Variational Dropout Variational Dropout is a regularization technique based on dropout, but uses a variational inference grounded approach. In…

Similar Papers 제목 키워드 기반

Breaking the Softmax Bottleneck via Learnable Monotonic Pointwise Non-linearities

2019-02-21 · Octavian-Eugen Ganea, Sylvain Gelly, Gary Bécigneul, Aliaksei Severyn

The Softmax function on top of a final linear layer is the de facto method to output probability distributions in neural networks. In many applications such as language models or text generation, this model has to produc…

Language ModelingLanguage ModellingText Generation

Sigsoftmax: Reanalysis of the Softmax Bottleneck

2018-05-28 · NeurIPS 2018 12 · Sekitoshi Kanai, Yasuhiro Fujiwara, Yuki Yamanaka, Shuichi Adachi

Softmax is an output activation function for modeling categorical probability distributions in many applications of deep learning. However, a recent study revealed that softmax can be a bottleneck of representational cap…

Language ModelingLanguage Modelling

Breaking the Softmax Bottleneck for Sequential Recommender Systems with Dropout and Decoupling

2021-10-11 · Ying-Chen Lin

The Softmax bottleneck was first identified in language modeling as a theoretical limit on the expressivity of Softmax-based models. Being one of the most widely-used methods to output probability, Softmax-based models h…

Language ModelingLanguage ModellingRecommendation Systems

Mixtape: Breaking the Softmax Bottleneck Efficiently

2019-12-01 · NeurIPS 2019 12 · Zhilin Yang, Thang Luong, Russ R. Salakhutdinov, Quoc V. Le

The softmax bottleneck has been shown to limit the expressiveness of neural language models. Mixture of Softmaxes (MoS) is an effective approach to address such a theoretical limitation, but are expensive compared to sof…

Language ModelingLanguage ModellingMachine TranslationTranslation+1

Why do small language models underperform? Studying Language Model Saturation via the Softmax Bottleneck

2024-04-11 · Nathan Godey, Éric de la Clergerie, Benoît Sagot

Recent advances in language modeling consist in pretraining highly parameterized neural networks on extremely large web-mined text corpora. Training and inference with such models can be costly in practice, which incenti…

Language ModelingLanguage Modelling