Due to recent pretrained multilingual representation models, it has become feasible to exploit labeled data from one language to train a cross-lingual model that can then be applied to multiple new languages.
Understanding the behavior and vulnerability of pre-trained deep neural networks (DNNs) can help to improve them.
Moreover, our calibration algorithm can produce SNN with state-of-the-art architecture on the large-scale ImageNet dataset, including MobileNet and RegNet.
We introduce Channel Mean Discrepancy (CMD), a model-agnostic distance metric for evaluating the statistics of features extracted by classification models, inspired by integral probability metrics.
no code implementations • 16 Feb 2021 • John Arrington, Reynier Cruz-Torres, Winston DeGraw, Xin Dong, Leo Greiner, Samuel Heppelmann, Barbara Jacak, Yuanjing Ji, Matthew Kelsey, Spencer R. Klein, Yue Shi Lai, Grazyna Odyniec, Sooraj Radhakrishnan, Ernst Sichtermann, Youqi Son, Fernando Torales Acosta, Lei Xia, Nu Xu, Feng Yuan, Yuxiang Zhao
The proposed electron-ion collider has a rich physics program to study the internal structure of protons and heavy nuclei.
Nuclear Experiment Nuclear Theory
Sentiment analysis is an area of substantial relevance both in industry and in academia, including for instance in social studies.
However, the inversion process only utilizes biased feature statistics stored in one model and is from low-dimension to high-dimension.
We introduce exBERT, a training method to extend BERT pre-trained models from a general domain to a new pre-trained model for a specific domain with a new additive vocabulary under constrained training resources (i. e., constrained computation and data).
The resulting model then serves as a teacher to induce labels for unlabeled target language samples that can be used during further adversarial training, allowing us to gradually adapt our model to the target language.
Network quantization has rapidly become one of the most widely used methods to compress and accelerate deep neural networks.
A number of cross-lingual transfer learning approaches based on neural networks have been proposed for the case when large amounts of parallel text are at our disposal.
In practice, however, these two sets of reviews are notably different: users' reviews reflect a variety of items that they have bought and are hence very heterogeneous in their topics, while an item's reviews pertain only to that single item and are thus topically homogeneous.
We first bring up three omitted issues in extremely low-bit networks: the squashing range of quantized values; the gradient vanishing during backpropagation and the unexploited hardware acceleration of ternary networks.
Based on massive amounts of data, recent pretrained contextual representation models have made significant strides in advancing a number of different English NLP tasks.
We propose Additive Powers-of-Two~(APoT) quantization, an efficient non-uniform quantization scheme for the bell-shaped and long-tailed distribution of weights and activations in neural networks.
A highlight of our full-stack approach which attributes to the achieved high energy efficiency is an efficient Selector-Accumulator (SAC) architecture for implementing the multiplier-accumulator (MAC) operation present in any digital CNN hardware.
To reduce memory footprint and run-time latency, techniques such as neural network pruning and binarization have been explored separately.
In this work, we propose a novel topic consisting of two dual tasks: 1) given a scene, recommend objects to insert, 2) given an object category, retrieve suitable background scenes.
Deep convolutional neural networks excel at sentiment polarity classification, but tend to require substantial amounts of training data, which moreover differs quite significantly between domains.
Ranked #61 on Sentiment Analysis on SST-2 Binary classification
Binary neural networks (BNN) have been studied extensively since they run dramatically faster at lower memory and power consumption than floating-point networks, thanks to the efficiency of bit operations.
We demonstrate that our pose-based framework can achieve better accuracy than the state-of-art detection-based approach on the human instance segmentation problem, and can moreover better handle occlusion.
Ranked #1 on Human Instance Segmentation on OCHuman
How to develop slim and accurate deep neural networks has become crucial for real- world applications, especially for those employed in embedded systems.