arXiv subjects

Xiangyang Li

Publications and source records attributed to Xiangyang Li.

76 records · Page 5Linked to original sources

arXiv2019

Multifaceted Analysis of Fine-Tuning in Deep Model for Visual Recognition

In recent years, convolutional neural networks (CNNs) have achieved impressive performance for various visual recognition scenarios. CNNs trained on large labeled datasets can not only obtain significant performance on most challenging benchmarks but also provide powerful representations, which can be used to a wide range of other tasks. However, the requirement of massive amounts of data to train deep neural networks is a major drawback of these models, as the data available is usually limited or imbalanced. Fine-tuning (FT) is an effective way to transfer knowledge learned in a source dataset to a target task. In this paper, we introduce and systematically investigate several factors that influence the performance of fine-tuning for visual recognition. These factors include parameters for the retraining procedure (e.g., the initial learning rate of fine-tuning), the distribution of the source and target data (e.g., the number of categories in the source dataset, the distance between the source and target datasets) and so on. We quantitatively and qualitatively analyze these factors, evaluate their influence, and present many empirical observations. The results reveal insights into what fine-tuning changes CNN parameters and provide useful and evidence-backed intuitions about how to implement fine-tuning for computer vision tasks.

cs.CV↗

arXiv2018

Scene recognition with CNNs: objects, scales and dataset bias

Since scenes are composed in part of objects, accurate recognition of scenes requires knowledge about both scenes and objects. In this paper we address two related problems: 1) scale induced dataset bias in multi-scale convolutional neural network (CNN) architectures, and 2) how to combine effectively scene-centric and object-centric knowledge (i.e. Places and ImageNet) in CNNs. An earlier attempt, Hybrid-CNN, showed that incorporating ImageNet did not help much. Here we propose an alternative method taking the scale into account, resulting in significant recognition gains. By analyzing the response of ImageNet-CNNs and Places-CNNs at different scales we find that both operate in different scale ranges, so using the same network for all the scales induces dataset bias resulting in limited performance. Thus, adapting the feature extractor to each particular scale (i.e. scale-specific CNNs) is crucial to improve recognition, since the objects in the scenes have their specific range of scales. Experimental results show that the recognition accuracy highly depends on the scale, and that simple yet carefully chosen multi-scale combinations of ImageNet-CNNs and Places-CNNs, can push the state-of-the-art recognition accuracy in SUN397 up to 66.26% (and even 70.17% with deeper architectures, comparable to human performance).

cs.CV↗

arXiv2013

Throughput Optimizing Localized Link Scheduling for Multihop Wireless Networks Under Physical Interference Model

We study throughput-optimum localized link scheduling in wireless networks. The majority of results on link scheduling assume binary interference models that simplify interference constraints in actual wireless communication. While the physical interference model reflects the physical reality more precisely, the problem becomes notoriously harder under the physical interference model. There have been just a few existing results on link scheduling under the physical interference model, and even fewer on more practical distributed or localized scheduling. In this paper, we tackle the challenges of localized link scheduling posed by the complex physical interference constraints. By cooperating the partition and shifting strategies into the pick-and-compare scheme, we present a class of localized scheduling algorithms with provable throughput guarantee subject to physical interference constraints. The algorithm in the linear power setting is the first localized algorithm that achieves at least a constant fraction of the optimal capacity region subject to physical interference constraints. The algorithm in the uniform power setting is the first localized algorithm with a logarithmic approximation ratio to the optimal solution. Our extensive simulation results demonstrate correctness and performance efficiency of our algorithms.

cs.NI↗

arXiv2012

Efficient and Secure Key Extraction using CSI without Chasing down Errors

Generating keys and keeping them secret is critical in secure communications. Due to the "open-air" nature, key distribution is more susceptible to attacks in wireless communications. An ingenious solution is to generate common secret keys by two communicating parties separately without the need of key exchange or distribution, and regenerate them on needs. Recently, it is promising to extract keys by measuring the random variation in wireless channels, e.g., RSS. In this paper, we propose an efficient Secret Key Extraction protocol without Chasing down Errors, SKECE. It establishes common cryptographic keys for two communicating parties in wireless networks via the realtime measurement of Channel State Information (CSI). It outperforms RSS-based approaches for key generation in terms of multiple subcarriers measurement, perfect symmetry in channel, rapid decorrelation with distance, and high sensitivity towards environments. In the SKECE design, we also propose effective mechanisms such as the adaptive key stream generation, leakage resilient consistence validation, and weighted key recombination, to fully exploit the excellent properties of CSI. We implement SKECE on off-the-shelf 802.11n devices and evaluate its performance via extensive experiments. The results demonstrate that SKECE achieves a more than 3x throughput gain in the key generation from one subcarrier in static scenarios, and due to its high efficiency, a 50% reduction on the communication overhead compared to the state-of-the-art RSS based approaches.

cs.CR↗