In this work, we present a systematic method to reduce the performance degradation and inference complexity of Quantized Transformers.
2 code implementations • 6 Oct 2021 • Victor Schmidt, Alexandra Sasha Luccioni, Mélisande Teng, Tianyu Zhang, Alexia Reynaud, Sunand Raghupathi, Gautier Cosne, Adrien Juraver, Vahe Vardanyan, Alex Hernandez-Garcia, Yoshua Bengio
Climate change is a major threat to humanity, and the actions required to prevent its catastrophic consequences include changes in both policy-making and individual behaviour.
In this work, we propose a novel deep hashing model with only a single learning objective.
Originally developed for imputing missing entries in low rank, or approximately low rank matrices, matrix completion has proven widely effective in many problems where there is no reason to assume low-dimensional linear structure in the underlying matrix, as would be imposed by rank constraints.
Recently, the Transformer module has been transplanted from natural language processing to computer vision.
This paper proposes a novel ternary hash encoding for learning to hash methods, which provides a principled more efficient coding scheme with performances better than those of the state-of-the-art binary hashing counterparts.
Recently, neural-symbolic architectures have achieved success on commonsense reasoning through effectively encoding relational structures retrieved from external knowledge graphs (KGs) and obtained state-of-the-art results in tasks such as (commonsense) question answering and natural language inference.
The main difficulty of person re-identification (ReID) lies in collecting annotated data and transferring the model across different domains.
An initial database made up with input and target data is first generated from the computational model, from which a processed database is deduced by conditioning the input data with respect to the target data using the nonparametric statistics.
Recently, knowledge graph (KG) augmented models have achieved noteworthy success on various commonsense reasoning tasks.
This paper investigates capabilities of Privacy-Preserving Deep Learning (PPDL) mechanisms against various forms of privacy attacks.
The fundamental difficulty in person re-identification (ReID) lies in learning the correspondence among individual cameras.
Ranked #14 on Unsupervised Domain Adaptation on Duke to Market
Quantization of weights of deep neural networks (DNN) has proven to be an effective solution for the purpose of implementing DNNs on edge devices such as mobiles, ASICs and FPGAs, because they have no sufficient resources to support computation involving millions of high precision weights and multiply-accumulate operations.
Vehicle re-identification (Re-ID) is very important in intelligent transportation and video surveillance. Prior works focus on extracting discriminative features from visual appearance of vehicles or using visual-spatio-temporal information. However, background interference in vehicle re-identification have not been explored. In the actual large-scale spatio-temporal scenes, the same vehicle usually appears in different backgrounds while different vehicles might appear in the same background, which will seriously affect the re-identification performance.
Differently, this paper investigates ReID in an unexplored single-camera-training (SCT) setting, where each person in the training set appears in only one camera.