EfficientCLIP: Efficient Cross-Modal Pre-training by Ensemble Confident Learning and Language Modeling

While large scale pre-training has achieved great achievements in bridging the gap between vision and language, it still faces several challenges. First, the cost for pre-training is expensive. Second, there is no efficient way to handle the data noise which degrades model performance. Third, previous methods only leverage limited image-text paired data, while ignoring richer single-modal data, which may result in poor generalization to single-modal downstream tasks. In this work, we propose an EfficientCLIP method via Ensemble Confident Learning to obtain a less noisy data subset. Extra rich non-paired single-modal text data is used for boosting the generalization of text branch. We achieve the state-of-the-art performance on Chinese cross-modal retrieval tasks with only 1/10 training resources compared to CLIP and WenLan, while showing excellent generalization to single-modal tasks, including text retrieval and text classification.

Joint Multi-User Communication and Sensing Exploiting Both Signal and Environment Sparsity

As a potential technology feature for 6G wireless networks, the idea of sensing-communication integration requires the system not only to complete reliable multi-user communication but also to achieve accurate environment sensing. In this paper, we consider such a joint communication and sensing (JCAS) scenario, in which multiple users use the sparse code multiple access (SCMA) scheme to communicate with the wireless access point (AP). Part of the user signals are scattered by the environment object and reflected by an intelligent reflective surface (IRS) before they arrive at the AP. We exploit the sparsity of both the structured user signals and the unstructured environment and propose an iterative and incremental joint multi-user communication and environment sensing scheme, in which the two processes, i.e., multi-user information detection and environment object detection, interweave with each other thanks to their intrinsic mutual dependence. The proposed algorithm is sliding-window based and also graph based, which can keep on sensing the environment as long as there are illuminating user signals. The trade-off relationship between the key system parameters is analyzed, and the simulation result validates the convergence and effectiveness of the proposed algorithm.

Generalized One-Class Learning Using Pairs of Complementary Classifiers

One-class learning is the classic problem of fitting a model to the data for which annotations are available only for a single class. In this paper, we explore novel objectives for one-class learning, which we collectively refer to as Generalized One-class Discriminative Subspaces (GODS). Our key idea is to learn a pair of complementary classifiers to flexibly bound the one-class data distribution, where the data belongs to the positive half-space of one of the classifiers in the complementary pair and to the negative half-space of the other. To avoid redundancy while allowing non-linearity in the classifier decision surfaces, we propose to design each classifier as an orthonormal frame and seek to learn these frames via jointly optimizing for two conflicting objectives, namely: i) to minimize the distance between the two frames, and ii) to maximize the margin between the frames and the data. The learned orthonormal frames will thus characterize a piecewise linear decision surface that allows for efficient inference, while our objectives seek to bound the data within a minimal volume that maximizes the decision margin, thereby robustly capturing the data distribution. We explore several variants of our formulation under different constraints on the constituent classifiers, including kernelized feature maps. We demonstrate the empirical benefits of our approach via experiments on data from several applications in computer vision, such as anomaly detection in video sequences, human poses, and human activities. We also explore the generality and effectiveness of GODS for non-vision tasks via experiments on several UCI datasets, demonstrating state-of-the-art results.

Long-Short Temporal Contrastive Learning of Video Transformers

Video transformers have recently emerged as a competitive alternative to 3D CNNs for video understanding. However, due to their large number of parameters and reduced inductive biases, these models require supervised pretraining on large-scale image datasets to achieve top performance. In this paper, we empirically demonstrate that self-supervised pretraining of video transformers on video-only datasets can lead to action recognition results that are on par or better than those obtained with supervised pretraining on large-scale image datasets, even massive ones such as ImageNet-21K. Since transformer-based models are effective at capturing dependencies over extended temporal spans, we propose a simple learning procedure that forces the model to match a long-term view to a short-term view of the same video. Our approach, named Long-Short Temporal Contrastive Learning (LSTCL), enables video transformers to learn an effective clip-level representation by predicting temporal context captured from a longer temporal extent. To demonstrate the generality of our findings, we implement and validate our approach under three different self-supervised contrastive learning frameworks (MoCo v3, BYOL, SimSiam) using two distinct video-transformer architectures, including an improved variant of the Swin Transformer augmented with space-time attention. We conduct a thorough ablation study and show that LSTCL achieves competitive performance on multiple video benchmarks and represents a convincing alternative to supervised image-based pretraining.

On Selection of Semiparametric Spatial Regression Models

In this paper, we focus on the variable selection techniques for a class of semiparametric spatial regression models which allow one to study the effects of explanatory variables in the presence of the spatial information. The spatial smoothing problem in the nonparametric part is tackled by means of bivariate splines over triangulation, which is able to deal efficiently with data distributed over irregularly shaped regions. In addition, we develop a unified procedure for variable selection to identify significant covariates under a double penalization framework, and we show that the penalized estimators enjoy the oracle property. The proposed method can simultaneously identify non-zero spatially distributed covariates and solve the problem of leakage across complex domains of the functional spatial component. To estimate the standard deviations of the proposed estimators for the coefficients, a sandwich formula is developed as well. In the end, Monte Carlo simulation examples and a real data example are provided to illustrate the proposed methodology. All technical proofs are given in the supplementary materials.

Physical Layer Security for UAV Communications in 5G and Beyond Networks

Due to its high mobility and flexible deployment, unmanned aerial vehicle (UAV) is drawing unprecedented interest in both military and civil applications to enable agile wireless communications and provide ubiquitous connectivity. Mainly operating in an open environment, UAV communications can benefit from dominant line-of-sight links; however, it on the other hand renders the UAVs more vulnerable to malicious eavesdropping or jamming attacks. Recently, physical layer security (PLS), which exploits the inherent randomness of the wireless channels for secure communications, has been introduced to UAV systems as an important complement to the conventional cryptography-based approaches. In this paper, a comprehensive survey on the current achievements of the UAV-aided wireless communications is conducted from the PLS perspective. We first introduce the basic concepts of UAV communications including the typical static/mobile deployment scenarios, the unique characteristics of air-to-ground channels, as well as various roles that a UAV may act when PLS is concerned. Then, we introduce the widely used secrecy performance metrics and start by reviewing the secrecy performance analysis and enhancing techniques for statically deployed UAV systems, and extend the discussion to a more general scenario where the UAVs mobility is further exploited. For both cases, respectively, we summarize the commonly adopted methodologies in the corresponding analysis and design, then describe important works in the literature in detail. Finally, potential research directions and challenges are discussed to provide an outlook for future works in the area of UAV-PLS in 5G and beyond networks.

Learning Log-Determinant Divergences for Positive Definite Matrices

Representations in the form of Symmetric Positive Definite (SPD) matrices have been popularized in a variety of visual learning applications due to their demonstrated ability to capture rich second-order statistics of visual data. There exist several similarity measures for comparing SPD matrices with documented benefits. However, selecting an appropriate measure for a given problem remains a challenge and in most cases, is the result of a trial-and-error process. In this paper, we propose to learn similarity measures in a data-driven manner. To this end, we capitalize on the alphabeta-log-det divergence, which is a meta-divergence parametrized by scalars alpha and beta, subsuming a wide family of popular information divergences on SPD matrices for distinct and discrete values of these parameters. Our key idea is to cast these parameters in a continuum and learn them from data. We systematically extend this idea to learn vector-valued parameters, thereby increasing the expressiveness of the underlying non-linear measure. We conjoin the divergence learning problem with several standard tasks in machine learning, including supervised discriminative dictionary learning and unsupervised SPD matrix clustering. We present Riemannian gradient descent schemes for optimizing our formulations efficiently, and show the usefulness of our method on eight standard computer vision tasks.

TAG: Transformer Attack from Gradient

Although federated learning has increasingly gained attention in terms of effectively utilizing local devices for data privacy enhancement, recent studies show that publicly shared gradients in the training process can reveal the private training images (gradient leakage) to a third-party in computer vision. We have, however, no systematic understanding of the gradient leakage mechanism on the Transformer based language models. In this paper, as the first attempt, we formulate the gradient attack problem on the Transformer-based language models and propose a gradient attack algorithm, TAG, to reconstruct the local training data. We develop a set of metrics to evaluate the effectiveness of the proposed attack algorithm quantitatively. Experimental results on Transformer, TinyBERT$_{4}$, TinyBERT$_{6}$, BERT$_{BASE}$, and BERT$_{LARGE}$ using GLUE benchmark show that TAG works well on more weight distributions in reconstructing training data and achieves 1.5$times$ recover rate and 2.5$times$ ROUGE-2 over prior methods without the need of ground truth label. TAG can obtain up to 90$%$ data by attacking gradients in CoLA dataset. In addition, TAG has a stronger adversary on large models, small dictionary size, and small input length. We hope the proposed TAG will shed some light on the privacy leakage problem in Transformer-based NLP models.

MetaSCI: Scalable and Adaptive Reconstruction for Video Compressive Sensing

To capture high-speed videos using a two-dimensional detector, video snapshot compressive imaging (SCI) is a promising system, where the video frames are coded by different masks and then compressed to a snapshot measurement. Following this, efficient algorithms are desired to reconstruct the high-speed frames, where the state-of-the-art results are achieved by deep learning networks. However, these networks are usually trained for specific small-scale masks and often have high demands of training time and GPU memory, which are hence {bf em not flexible} to $i$) a new mask with the same size and $ii$) a larger-scale mask. We address these challenges by developing a Meta Modulated Convolutional Network for SCI reconstruction, dubbed MetaSCI. MetaSCI is composed of a shared backbone for different masks, and light-weight meta-modulation parameters to evolve to different modulation parameters for each mask, thus having the properties of {bf em fast adaptation} to new masks (or systems) and ready to {bf em scale to large data}. Extensive simulation and real data results demonstrate the superior performance of our proposed approach. Our code is available at {smallurl{https://github.com/xyvirtualgroup/MetaSCI-CVPR2021}}.

VX2TEXT: End-to-End Learning of Video-Based Text Generation From Multimodal Inputs

We present textsc{Vx2Text}, a framework for text generation from multimodal inputs consisting of video plus text, speech, or audio. In order to leverage transformer networks, which have been shown to be effective at modeling language, each modality is first converted into a set of language embeddings by a learnable tokenizer. This allows our approach to perform multimodal fusion in the language space, thus eliminating the need for ad-hoc cross-modal fusion modules. To address the non-differentiability of tokenization on continuous inputs (e.g., video or audio), we utilize a relaxation scheme that enables end-to-end training. Furthermore, unlike prior encoder-only models, our network includes an autoregressive decoder to generate open-ended text from the multimodal embeddings fused by the language encoder. This renders our approach fully generative and makes it directly applicable to different video+$x$ to text problems without the need to design specialized network heads for each task. The proposed framework is not only conceptually simple but also remarkably effective: experiments demonstrate that our approach based on a single architecture outperforms the state-of-the-art on three video-based text-generation tasks -- captioning, question answering and audio-visual scene-aware dialog.