Recently, many self-supervised pre-training methods have been proposed to improve the performance of deep neural networks (DNNs) for 3D point clouds processing. However, the common mechanism underlying the effectiveness of different pre-training methods remains unclear. In this paper, we use game-theoretic interactions as a unified approach to explore the common mechanism of pre-training methods. Specifically, we decompose the output score of a DNN into the sum of numerous effects of interactions, with each interaction representing a distinct 3D substructure of the input point cloud. Based on the decomposed interactions, we draw the following conclusions. (1) The common mechanism across different pre-training methods is that they enhance the strength of high-order interactions encoded by DNNs, which represent complex and global 3D structures, while reducing the strength of low-order interactions, which represent simple and local 3D structures. (2) Sufficient pre-training and adequate fine-tuning data for downstream tasks further reinforce the mechanism described above. (3) Pre-training methods carry a potential risk of reducing the transferability of features encoded by DNNs. Inspired by the observed common mechanism, we propose a new method to directly enhance the strength of high-order interactions and reduce the strength of low-order interactions encoded by DNNs, improving performance without the need for pre-training on large-scale datasets. Experiments show that our method achieves performance comparable to traditional pre-training methods.

By adding exiting layers to the deep learning networks, early exit can terminate the inference earlier with accurate results. However, the passive decision-making of whether to exit or continue the next layer has to go through every pre-placed exiting layer until it exits. In addition, it is hard to adjust the configurations of the computing platforms alongside the inference proceeds. By incorporating a low-cost prediction engine, we propose a Predictive Exit framework for computation- and energy-efficient deep learning applications. Predictive Exit can forecast where the network will exit (i.e., establish the number of remaining layers to finish the inference), which effectively reduces the network computation cost by exiting on time without running every pre-placed exiting layer. Moreover, according to the number of remaining layers, proper computing configurations (i.e., frequency and voltage) are selected to execute the network to further save energy. Extensive experimental results demonstrate that Predictive Exit achieves up to 96.2% computation reduction and 72.9% energy-saving compared with classic deep learning networks; and 12.8% computation reduction and 37.6% energy-saving compared with the early exit under state-of-the-art exiting strategies, given the same inference accuracy and latency.

360° videos are becoming one of the major media in recent years, providing immersive experience for viewers with more interactions compared to traditional videos. Most of today’s implementations rely on bulky Head-Mounted Displays (HMDs) or require touch screen operations for interactive display, which are not only expensive but also inconvenient for viewers. In this paper, we demonstrate that interactive 360° video streaming can be done with hints from gaze movement detected by the front camera of today’s mobile devices (e.g., a smartphone). We design a lightweight real-time gaze point tracking method for this purpose. Using only the front camera, our solution detects the users’ faces by a lightweight Haar-like cascaded classifier, measures the user’s face-to-screen distance and sight angle, and then derives the location of the user’s gaze point following a customized triangularity model. We integrate it with streaming module and apply a dynamic margin adaption algorithm to minimize the overall energy consumption for battery-constrained mobile devices. Our experiments on state-of-the-art smartphones show the feasibility of our solution and its energy efficiency toward cost-effective real-time 360° video streaming.

Tile-based approach is widely adopted in adaptive 360\textdegree~video streaming systems. Existing QoE-driven streaming approaches usually obtain the tile selection and adjust the bitrate based on the viewport prediction with a fixed tiling, which fail to consider the unstable prediction performance. However, varying the tiling of the video can produce different number of tiles with different sizes, and thus can have distinct impacts on error tolerance for viewport prediction and on decoding complexity for resource-constrained mobile client. In this work, we introduce adaptive tiling into the conventional bitrate adaptation for mobile 360degree~video streaming. We first analyze the impacts of tilings on tile selection and decoding time, which verify the benefit of tiling adaptation in various practical aspects. We then formulate the QoE optimization problem for adaptive tiling and bitrate streaming and discuss the design details of our adaptation algorithm, which can adapt to the performance of viewport prediction and the decoding capabilities of mobile clients in addition to the conventional influencing factors. Finally, the superiority of our proposed approach compared with the state-of-the-art methods is evaluated through extensive trace-driven simulations.

Cloud computing platforms enable applications to offer low-latency services to users by deploying data storage in multiple geo-distributed data centers. In this paper, through benchmark measurements on Amazon AWS and Microsoft Azure together with an analysis of a large-scale dataset collected from a major cloud CDN provider, we identify the high tail latency problem in cloud CDNs, which can substantially undermine the efficacy of cloud CDNs. One crucial idea to reduce the tail latency is to send requests in parallel to multiple clouds in cloud CDNs. However, since application providers often have a budget for using cloud services, deciding how many chunks to download from each cloud and when to download chunks in a cost-efficient manner still remain as open problems in our concerned scenario. To address the problem, we present TailCutter, a workload scheduling framework that aims at optimizing the tail latency while meeting cost constraints given by application providers. Specifically, we formulate the tail latency minimization (TLM) problem in cloud CDNs and design the receding horizon control based maximum tail minimization algorithm (RHC-based MTMA) to efficiently solve the TLM problem in practice. We implement TailCutter across multiple data centers of Amazon AWS and Microsoft Azure. Extensive evaluations using a large-scale real-world data trace (collected from a major ISP) illustrate that TailCutter can reduce up to 58.9% of the 100th-percentile user-perceived latency, as compared with alternative solutions under the cost constraint.

Recent years have seen various acoustic applications on mobile devices, e.g. range finding, gesture recognition, and device-to-device data transport, which use near-ultrasound signals at frequencies around 18-24 kHz. Due to the fixed low sound sample rate and hardware limitation, the highest detectable sound frequency on commercial-off-the-shelf (COTS) mobile devices is capped at 24 kHz, presenting a daunting barrier that prevents high-frequency ultrasounds from benefiting acoustic applications. To bridge this gap, we present iChemo, a technology that enables COTS mobile devices to sense high-frequency ultrasound signals. Specifically, we demonstrate how to detect the power spectral density (PSD) of a high-frequency ultrasound signal by customizing the coprime sampling algorithm on COTS devices. Through our prototype and evaluation on extensive mobile devices, we demonstrate that iChemo can sense the PSD of ultrasound at frequency of 60 kHz, which is over twice of the current sensible frequency threshold.

Many efforts have been made on sensing ultrasound with commercial-off-the-shelf (COTS) mobile devices in the recent literature. Yet due to the limited sound sample rate, current COTS mobile devices can not directly capture any sound at the frequency over 24 kHz. This issue prevents true ultrasound, of which the frequency is typically over 40 kHz, from benefiting the existing sound sensing applications. In this work, we show that by subtly customizing the sampling process, we can make COTS mobile devices hear the true ultrasound that is typically beyond their capability to fully capture. Particularly, we present a system that enable COTS mobile devices to sense the power spectrum density (PSD) of true ultrasounds, of which the frequency can be as high as 60 kHz.

Ensuring high video quality of experience (QoE) on the broadcaster side is critical for interactive live streaming. However, measurements on multiple live streaming platforms show that they all suffer from broadcaster-side video quality degradation in the presence of transient bandwidth fluctuations. This paper presents Greedy Variable Bitrate (GVBR), a suite of solutions that optimizes the QoE through an approriate keyframe interval that trades cross-frame compression for lowered inter-frame interdependency, a simple-yet-efficient frame dropping strategy to prevent excessive frame drops, and a bitrate adaptation strategy customized for broadcasters who have shallow buffer. We compare GVBR with state-of-art algorithms in different network conditions, and find that GVBR can cut video interruption incidents by 90%, while achieving comparable bitrate.

Energy consumption has become a key issue for running large-scale data center networks (DCN) nowadays. Previous studies mainly focus on energy saving through reducing the number of active servers or network switches with traffic consolidation. However, since this mechanism benefits from the routing flexibility, both the gained energy savings and flow performance are limited by the conventional static network topology. Recent advances in DCN architecture propose to implement an adaptive network topology with reconfigurable optical/wireless links (i.e., topology-adaptive DCNs), which has shown a great potential to improve the transmission performance of existing static wired network. In this paper, we propose GreenWay, an energy-efficient solution to jointly optimize the virtual machine (VM) placement and flow transmissions under the new paradigm of topology adaption. Based on the VM traffic demands, it can construct a proper run-time network topology to lower both the energy cost and communication cost among VMs. We first formulate the energy optimization problem in topology-adaptive DCNs. After showing the its NP-hardness, we then develop heuristic algorithms to address the VM placement and topology adaption effectively. Extensive trace-based simulations show that GreenWay consumes much less energy than other state-of-the- art solutions while ensuring better flow performance. We finally implement an OpenvSwitch- based testbed and demonstrate the efficiency of our solution.

Novel cloud computing technology makes multiplayer cloud gaming a reality, where players play games that do not run on local devices, but on servers in the cloud. Nevertheless, the necessary communication between player's local device and cloud server increases the response delay of the gaming session. Besides the degrade of responsiveness, the inter-player delay, which is the difference of response delays perceived by players who are interacting with each other, can significantly affect the fairness of the multiplayer game. In this paper, we first introduce the Inter-player Delay Optimization (IDO) problem that aims at minimizing this inter-player delay, while preserving good-enough absolute response delay experienced by players. We further propose an efficient heuristic algorithm to solve the IDO problem. The evaluation result of a comprehensive simulation using a large-scale real-world data trace shows that IDO can reduce up to about 30% of maximum inter-player delay among interacting players comparing with the other existing solution.

Cloud computing platforms enable applications to offer low latency access to user data by offering storage services in several geographically distributed data centers. In this paper, we identify the high tail latency problem in cloud CDN via analyzing a large-scale dataset collected from 783,944 users in a major cloud CDN. We find that the data downloading latency in cloud CDN is highly variable, which may significantly degrade the user experience of applications. To address the problem, we present TailCutter, a workload scheduling mechanism that aims at optimizing the tail latency while meeting the cost constraint given by application providers. We further design the Maximum Tail Minimization Algorithm (MTMA) working in TailCutter mechanism to optimally solve the Tail Latency Minimization (TLM) problem in polynomial time. We implement TailCutter across data centers of Amazon S3 and Microsoft Azure. Our extensive evaluation using large-scale real world data traces shows that TailCutter can reduce up to 68% 99th percentile user-perceived latency in comparison with alternative solutions under cost constraints.