In this work, we connect two distinct concepts for unsupervised domain adaptation: feature distribution alignment between domains by utilizing the task-specific decision boundary and the Wasserstein metric. Our proposed sliced Wasserstein discrepancy (SWD) is designed to capture the natural notion of dissimilarity between the outputs of task-specific classifiers. It provides a geometrically meaningful guidance to detect target samples that are far from the support of the source and enables efficient distribution alignment in an end-to-end trainable fashion. In the experiments, we validate the effectiveness and genericness of our method on digit and sign recognition, image classification, semantic segmentation, and object detection.

/pdf/sliced-wasserstein-discrepancy-for-unsupervised-domain-vb1b0l6tka.pdf

Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptation

This work tackles the problem of semi-supervised learning of image classifiers. Our main insight is that the field of semi-supervised learning can benefit from the quickly advancing field of self-supervised visual representation learning. Unifying these two approaches, we propose the framework of self-supervised semi-supervised learning and use it to derive two novel semi-supervised image classification methods. We demonstrate the effectiveness of these methods in comparison to both carefully tuned baselines, and existing semi-supervised learning methods. We then show that our approach and existing semi-supervised methods can be jointly trained, yielding a new state-of-the-art result on semi-supervised ILSVRC-2012 with 10% of labels.

S4L: Self-Supervised Semi-Supervised Learning

Recent advances in domain adaptation show that deep self-training presents a powerful means for unsupervised domain adaptation. These methods often involve an iterative process of predicting on target domain and then taking the confident predictions as pseudo-labels for retraining. However, since pseudo-labels can be noisy, self-training can put overconfident label belief on wrong classes, leading to deviated solutions with propagated errors. To address the problem, we propose a confidence regularized self-training (CRST) framework, formulated as regularized self-training. Our method treats pseudo-labels as continuous latent variables jointly optimized via alternating optimization. We propose two types of confidence regularization: label regularization (LR) and model regularization (MR). CRST-LR generates soft pseudo-labels while CRST-MR encourages the smoothness on network output. Extensive experiments on image classification and semantic segmentation show that CRSTs outperform their non-regularized counterpart with state-of-the-art performance. The code and models of this work are available at https://github.com/yzou2/CRST.

/pdf/confidence-regularized-self-training-1mfa8thlzr.pdf

Confidence Regularized Self-Training

With the high demand for wireless data traffic, WiFi networks have experienced very rapid growth, because they provide high throughput and are easy to deploy. Recently, Channel State Information (CSI) measured by WiFi networks is widely used for different sensing purposes. To get a better understanding of existing WiFi sensing technologies and future WiFi sensing trends, this survey gives a comprehensive review of the signal processing techniques, algorithms, applications, and performance results of WiFi sensing with CSI. Different WiFi sensing algorithms and signal processing techniques have their own advantages and limitations and are suitable for different WiFi sensing applications. The survey groups CSI-based WiFi sensing applications into three categories, detection, recognition, and estimation, depending on whether the outputs are binary/multi-class classifications or numerical values. With the development and deployment of new WiFi technologies, there will be more WiFi sensing opportunities wherein the targets may go beyond from humans to environments, animals, and objects. The survey highlights three challenges for WiFi sensing: robustness and generalization, privacy and security, and coexistence of WiFi sensing and networking. Finally, the survey presents three future WiFi sensing trends, i.e., integrating cross-layer network information, multi-device cooperation, and fusion of different sensors, for enhancing existing WiFi sensing capabilities and enabling new WiFi sensing opportunities.

WiFi Sensing with Channel State Information: A Survey

Domain adaptation refers to the problem of leveraging labeled data in a source domain to learn an accurate model in a target domain where labels are scarce or unavailable. A recent approach for finding a common representation of the two domains is via domain adversarial training (Ganin & Lempitsky, 2015), which attempts to induce a feature extractor that matches the source and target feature distributions in some feature space. However, domain adversarial training faces two critical limitations: 1) if the feature extraction function has high-capacity, then feature distribution matching is a weak constraint, 2) in non-conservative domain adaptation (where no single classifier can perform well in both the source and target domains), training the model to do well on the source domain hurts performance on the target domain. In this paper, we address these issues through the lens of the cluster assumption, i.e., decision boundaries should not cross high-density data regions. We propose two novel and related models: 1) the Virtual Adversarial Domain Adaptation (VADA) model, which combines domain adversarial training with a penalty term that punishes the violation the cluster assumption; 2) the Decision-boundary Iterative Refinement Training with a Teacher (DIRT-T) model, which takes the VADA model as initialization and employs natural gradient steps to further minimize the cluster assumption violation. Extensive empirical results demonstrate that the combination of these two models significantly improve the state-of-the-art performance on the digit, traffic sign, and Wi-Fi recognition domain adaptation benchmarks.

/pdf/a-dirt-t-approach-to-unsupervised-domain-adaptation-42nvg6s5se.pdf

A DIRT-T Approach to Unsupervised Domain Adaptation

In this article, we present a survey of recent advances in passive human behavior recognition in indoor areas using the channel state information (CSI) of commercial WiFi systems. The movement of the human body parts cause changes in the wireless signal reflections, which result in variations in the CSI. By analyzing the data streams of CSIs for different activities and comparing them against stored models, human behavior can be recognized. This is done by extracting features from CSI data streams and using machine learning techniques to build models and classifiers. The techniques from the literature that are presented herein have great performance; however, instead of the machine learning techniques employed in these works, we propose to use deep learning techniques such as long-short term memory (LSTM) recurrent neural networking (RNN) and show the improved performance. We also discuss different challenges such as environment change, frame rate selection, and the multi-user scenario; and finally suggest possible directions for future work.

A Survey on Behavior Recognition Using WiFi Channel State Information

Domain adaptation refers to the problem of leveraging labeled data in a source domain to learn an accurate model in a target domain where labels are scarce or unavailable A recent approach for finding a common representation of the two domains is via domain adversarial training (Ganin & Lempitsky, 2015), which attempts to induce a feature extractor that matches the source and target feature distributions in some feature space However, domain adversarial training faces two critical limitations: 1) if the feature extraction function has high-capacity, then feature distribution matching is a weak constraint, 2) in non-conservative domain adaptation (where no single classifier can perform well in both the source and target domains), training the model to do well on the source domain hurts performance on the target domain In this paper, we address these issues through the lens of the cluster assumption, ie, decision boundaries should not cross high-density data regions We propose two novel and related models: 1) the Virtual Adversarial Domain Adaptation (VADA) model, which combines domain adversarial training with a penalty term that punishes the violation the cluster assumption; 2) the Decision-boundary Iterative Refinement Training with a Teacher (DIRT-T) model, which takes the VADA model as initialization and employs natural gradient steps to further minimize the cluster assumption violation Extensive empirical results demonstrate that the combination of these two models significantly improve the state-of-the-art performance on the digit, traffic sign, and Wi-Fi recognition domain adaptation benchmarks

To achieve a nearly zero-delay operating point in a polarization-maintaining (PM) fiber Sagnac interferometer, two identical PM fibers were incorporated so that their two main axes were orthogonally coupled to each other. A simple fiber vibration sensor system was constructed with a light emitting diode and a balanced PM fiber Sagnac interferometer, in which one of the PM fibers was used as a sensing cable and the other as a reference cable. The vibration sensor was confirmed to be temperature-compensated and generated a phase shift per unit length and unit strain of the sensor of 4.7 milliradian/(m·μe) when mechanical vibrations with 1 kHz sinusoidal and triangular waves were stably observed under an input power of 10 μW.

Balanced polarization maintaining fiber Sagnac interferometer vibration sensor

In this article, we present a survey of recent advances in passive human behaviour recognition in indoor areas using the channel state information (CSI) of commercial WiFi systems. Movement of human body causes a change in the wireless signal reflections, which results in variations in the CSI. By analyzing the data streams of CSIs for different activities and comparing them against stored models, human behaviour can be recognized. This is done by extracting features from CSI data streams and using machine learning techniques to build models and classifiers. The techniques from the literature that are presented herein have great performances, however, instead of the machine learning techniques employed in these works, we propose to use deep learning techniques such as long-short term memory (LSTM) recurrent neural network (RNN), and show the improved performance. We also discuss about different challenges such as environment change, frame rate selection, and multi-user scenario, and suggest possible directions for future work.

Hirokazu Narui

Papers

A DIRT-T Approach to Unsupervised Domain Adaptation

A Survey on Behavior Recognition Using WiFi Channel State Information

A DIRT-T Approach to Unsupervised Domain Adaptation

Balanced polarization maintaining fiber Sagnac interferometer vibration sensor

A Survey of Human Activity Recognition Using WiFi CSI.