Search

Content addressable memory (CAM) stands out as an efficient hardware solution for memory-intensive search operations by supporting parallel computation in memory. However, developing a CAM-based accelerator architecture that achieves acceptable accuracy, while minimizing hardware cost and catering to both exact and approximate search, still presents a significant challenge especially when considering a broader spectrum of applications. This complexity stems from CAM's rapid evolution across multiple levels--algorithms, architectures, circuits, and underlying devices. This paper introduces CAMASim, a first comprehensive CAM accelerator simulation framework, emphasizing modularity, flexibility, and generality. CAMASim establishes the detailed design space for CAM-based accelerators, incorporates automated functional simulation for accuracy, and enables hardware performance prediction, by leveraging a circuit-level CAM modeling tool. This work streamlines the design space exploration for CAM-based accelerator, aiding researchers in developing effective CAM-based accelerators for various search-intensive applications.

Trusted I/O (TIO) is an appealing solution to improve I/O performance for confidential VMs (CVMs), with the potential to eliminate broad sources of I/O overhead. However, this paper emphasizes that not all types of I/O can derive substantial benefits from TIO, particularly network I/O. Given the obligatory use of encryption protocols for network traffic in CVM's threat model, TIO's approach of I/O encryption over the PCIe bus becomes redundant. Furthermore, TIO solutions need to expand the Trusted Computing Base (TCB) to include TIO devices and are commercially unavailable. Motivated by these insights, the goal of this paper is to propose a software solution that helps CVMs immediately benefit from high-performance networks, while confining trust only to the on-chip CVM. We present FOLIO, a software solution crafted from a secure and efficient Data Plane Development Kit (DPDK) extension compatible with the latest version of AMD Secure Encrypted Virtualization (SEV), a.k.a., Secure Nested Paging (SNP). Our design is informed by a thorough analysis of all possible factors that impact SNP VM's network performance. By extensively removing overhead sources, we arrive at a design that approaches the efficiency of an optimal TIO-based configuration. Evaluation shows that FOLIO has a performance dip less than 6% relative to the optimal TIO configuration, while only relying on off-the-shelf CPUs.

In this paper, we unveil a fundamental side channel in Wi-Fi networks, specifically the observable frame size, which can be exploited by attackers to conduct TCP hijacking attacks. Despite the various security mechanisms (e.g., WEP and WPA2/WPA3) implemented to safeguard Wi-Fi networks, our study reveals that an off path attacker can still extract sufficient information from the frame size side channel to hijack the victim's TCP connection. Our side channel attack is based on two significant findings: (i) response packets (e.g., ACK and RST) generated by TCP receivers vary in size, and (ii) the encrypted frames containing these response packets have consistent and distinguishable sizes. By observing the size of the victim's encrypted frames, the attacker can detect and hijack the victim's TCP connections. We validate the effectiveness of this side channel attack through two case studies, i.e., SSH DoS and web traffic manipulation. Precisely, our attack can terminate the victim's SSH session in 19 seconds and inject malicious data into the victim's web traffic within 28 seconds. Furthermore, we conduct extensive measurements to evaluate the impact of our attack on real-world Wi-Fi networks. We test 30 popular wireless routers from 9 well-known vendors, and none of these routers can protect victims from our attack. Besides, we implement our attack in 80 real-world Wi-Fi networks and successfully hijack the victim's TCP connections in 75 (93.75%) evaluated Wi-Fi networks. We have responsibly disclosed the vulnerability to the Wi-Fi Alliance and proposed several mitigation strategies to address this issue.

Machine learning and data analytics applications increasingly suffer from the high latency and energy consumption of conventional von Neumann architectures. Recently, several in-memory and near-memory systems have been proposed to remove this von Neumann bottleneck. Platforms based on content-addressable memories (CAMs) are particularly interesting due to their efficient support for the search-based operations that form the foundation for many applications, including K-nearest neighbors (KNN), high-dimensional computing (HDC), recommender systems, and one-shot learning among others. Today, these platforms are designed by hand and can only be programmed with low-level code, accessible only to hardware experts. In this paper, we introduce C4CAM, the first compiler framework to quickly explore CAM configurations and to seamlessly generate code from high-level TorchScript code. C4CAM employs a hierarchy of abstractions that progressively lowers programs, allowing code transformations at the most suitable abstraction level. Depending on the type and technology, CAM arrays exhibit varying latencies and power profiles. Our framework allows analyzing the impact of such differences in terms of system-level performance and energy consumption, and thus supports designers in selecting appropriate designs for a given application., Comment: 10 pages, 9 figures

Lattice-based cryptographic algorithms built on ring learning with error theory are gaining importance due to their potential for providing post-quantum security. However, these algorithms involve complex polynomial operations, such as polynomial modular multiplication (PMM), which is the most time-consuming part of these algorithms. Accelerating PMM is crucial to make lattice-based cryptographic algorithms widely adopted by more applications. This work introduces a novel high-throughput and compact PMM accelerator, X-Poly, based on the crossbar (XB)-type compute-in-memory (CIM). We identify the most appropriate PMM algorithm for XB-CIM. We then propose a novel bit-mapping technique to reduce the area and energy of the XB-CIM fabric, and conduct processing engine (PE)-level optimization to increase memory utilization and support different problem sizes with a fixed number of XB arrays. X-Poly design achieves 3.1X10^6 PMM operations/s throughput and offers 200X latency improvement compared to the CPU-based implementation. It also achieves 3.9X throughput per area improvement compared with the state-of-the-art CIM accelerators., Comment: Accepted by 42nd International Conference on Computer-Aided Design (ICCAD)

In this paper we propose a mask-conditional synthetic image generation model for creating synthetic satellite imagery datasets. Given a dataset of real high-resolution images and accompanying land cover masks, we show that it is possible to train an upstream conditional synthetic imagery generator, use that generator to create synthetic imagery with the land cover masks, then train a downstream model on the synthetic imagery and land cover masks that achieves similar test performance to a model that was trained with the real imagery. Further, we find that incorporating a mixture of real and synthetic imagery acts as a data augmentation method, producing better models than using only real imagery (0.5834 vs. 0.5235 mIoU). Finally, we find that encouraging diversity of outputs in the upstream model is a necessary component for improved downstream task performance. We have released code for reproducing our work on GitHub, see https://github.com/ms-synthetic-satellite-image/synthetic-satellite-imagery .

Image super-resolution is a common task on mobile and IoT devices, where one often needs to upscale and enhance low-resolution images and video frames. While numerous solutions have been proposed for this problem in the past, they are usually not compatible with low-power mobile NPUs having many computational and memory constraints. In this Mobile AI challenge, we address this problem and propose the participants to design an efficient quantized image super-resolution solution that can demonstrate a real-time performance on mobile NPUs. The participants were provided with the DIV2K dataset and trained INT8 models to do a high-quality 3X image upscaling. The runtime of all models was evaluated on the Synaptics VS680 Smart Home board with a dedicated edge NPU capable of accelerating quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 60 FPS rate when reconstructing Full HD resolution images. A detailed description of all models developed in the challenge is provided in this paper., Comment: arXiv admin note: text overlap with arXiv:2105.07825, arXiv:2105.08826, arXiv:2211.04470, arXiv:2211.03885, arXiv:2211.05256

Experience replay is an essential component in deep reinforcement learning (DRL), which stores the experiences and generates experiences for the agent to learn in real time. Recently, prioritized experience replay (PER) has been proven to be powerful and widely deployed in DRL agents. However, implementing PER on traditional CPU or GPU architectures incurs significant latency overhead due to its frequent and irregular memory accesses. This paper proposes a hardware-software co-design approach to design an associative memory (AM) based PER, AMPER, with an AM-friendly priority sampling operation. AMPER replaces the widely-used time-costly tree-traversal-based priority sampling in PER while preserving the learning performance. Further, we design an in-memory computing hardware architecture based on AM to support AMPER by leveraging parallel in-memory search operations. AMPER shows comparable learning performance while achieving 55x to 270x latency improvement when running on the proposed hardware compared to the state-of-the-art PER running on GPU., Comment: 9 pages, 9 figures. The work was accepted by the 41st International Conference on Computer-Aided Design (ICCAD), 2022, San Diego

Although the U-Net architecture has been extensively used for segmentation of medical images, we address two of its shortcomings in this work. Firstly, the accuracy of vanilla U-Net degrades when the target regions for segmentation exhibit significant variations in shape and size. Even though the U-Net already possesses some capability to analyze features at various scales, we propose to explicitly add multi-scale feature maps in each convolutional module of the U-Net encoder to improve segmentation of histology images. Secondly, the accuracy of a U-Net model also suffers when the annotations for supervised learning are noisy or incomplete. This can happen due to the inherent difficulty for a human expert to identify and delineate all instances of specific pathology very precisely and accurately. We address this challenge by introducing auxiliary confidence maps that emphasize less on the boundaries of the given target regions. Further, we utilize the bootstrapping properties of the deep network to address the missing annotation problem intelligently. In our experiments on a private dataset of breast cancer lymph nodes, where the primary task was to segment germinal centres and sinus histiocytosis, we observed substantial improvement over a U-Net baseline based on the two proposed augmentations., Comment: 12 pages, 4 figures , 2 tables ,Added Attention UNet Results, Added Sinus and Germinal Center overlay images, Modified paper format, Fixed Title typos

It is worth noticing that a fuzzy conjunction and its corresponding fuzzy implication can form a residual pair if and only if it is left-continuous. In order to get a more general result related on residual implications that induced by aggregation functions, we relax the definition of general overlap functions, more precisely, removing its right-continuous, and then introduce a new kind of aggregation functions, which called semi-overlap functions. Subsequently, we study some of their related algebraic properties and investigate their corresponding residual implications. Moreover, serval scholars have provided kinds of methods for fuzzy modus ponens (FMP,for short) problems so far, such as Zadeh's compositional rule of inference (CRI, for short), Wang's triple I method (TIM, for short) and the quintuple implication principle (QIP, for short). Compared with CRI and TIM methods, QIP method has some advantages in solving FMP problems, in this paper, we further consider the QIP method for FMP problems and prove that it satisfies the reducibility of multiple-rules fuzzy reasoning. Finally, we propose a new classification algorithm that based on semi-overlap functions and QIP method, which called SO5I-FRC algorithm. Through the comparative tests, the average accuracy of SO5I-FRC algorithm is higher than FARC-HD algorithm. The experimental results indicate that semi-overlap functions and QIP method have certain advantages and a wide range of applications in classification problems.

Recommendation systems (RecSys) suggest items to users by predicting their preferences based on historical data. Typical RecSys handle large embedding tables and many embedding table related operations. The memory size and bandwidth of the conventional computer architecture restrict the performance of RecSys. This work proposes an in-memory-computing (IMC) architecture (iMARS) for accelerating the filtering and ranking stages of deep neural network-based RecSys. iMARS leverages IMC-friendly embedding tables implemented inside a ferroelectric FET based IMC fabric. Circuit-level and system-level evaluation show that \fw achieves 16.8x (713x) end-to-end latency (energy) improvement compared to the GPU counterpart for the MovieLens dataset., Comment: Accepted by 59th Design Automation Conference (DAC)

Automated Guided Vehicles (AGVs) have been widely used for material handling in flexible shop floors. Each product requires various raw materials to complete the assembly in production process. AGVs are used to realize the automatic handling of raw materials in different locations. Efficient AGVs task allocation strategy can reduce transportation costs and improve distribution efficiency. However, the traditional centralized approaches make high demands on the control center's computing power and real-time capability. In this paper, we present decentralized solutions to achieve flexible and self-organized AGVs task allocation. In particular, we propose two improved multi-agent reinforcement learning algorithms, MADDPG-IPF (Information Potential Field) and BiCNet-IPF, to realize the coordination among AGVs adapting to different scenarios. To address the reward-sparsity issue, we propose a reward shaping strategy based on information potential field, which provides stepwise rewards and implicitly guides the AGVs to different material targets. We conduct experiments under different settings (3 AGVs and 6 AGVs), and the experiment results indicate that, compared with baseline methods, our work obtains up to 47\% task response improvement and 22\% training iterations reduction.

AMD's Secure Encrypted Virtualization (SEV) is an emerging security feature on AMD processors that allows virtual machines to run on encrypted memory and perform confidential computing even with an untrusted hypervisor. This paper first demystifies SEV's improper use of address space identifier (ASID) for controlling accesses of a VM to encrypted memory pages, cache lines, and TLB entries. We then present the CROSSLINE attacks, a novel class of attacks against SEV that allow the adversary to launch an attacker VM and change its ASID to that of the victim VM to impersonate the victim. We present two variants of CROSSLINE attacks: CROSSLINE V1 decrypts victim's page tables or memory blocks following the format of a page table entry; CROSSLINE V2 constructs encryption and decryption oracles by executing instructions of the victim VM. We have successfully performed CROSSLINE attacks on SEV and SEV-ES processors., Comment: 14 pages, 5 figures, security

Recently, BERT has become an essential ingredient of various NLP deep models due to its effectiveness and universal-usability. However, the online deployment of BERT is often blocked by its large-scale parameters and high computational cost. There are plenty of studies showing that the knowledge distillation is efficient in transferring the knowledge from BERT into the model with a smaller size of parameters. Nevertheless, current BERT distillation approaches mainly focus on task-specified distillation, such methodologies lead to the loss of the general semantic knowledge of BERT for universal-usability. In this paper, we propose a sentence representation approximating oriented distillation framework that can distill the pre-trained BERT into a simple LSTM based model without specifying tasks. Consistent with BERT, our distilled model is able to perform transfer learning via fine-tuning to adapt to any sentence-level downstream task. Besides, our model can further cooperate with task-specific distillation procedures. The experimental results on multiple NLP tasks from the GLUE benchmark show that our approach outperforms other task-specific distillation methods or even much larger models, i.e., ELMO, with efficiency well-improved.

This paper proposes a model-driven deep learning-based downlink channel reconstruction scheme for frequency division duplexing (FDD) massive multi-input multi-output (MIMO) systems. The spatial non-stationarity, which is the key feature of the future extremely large aperture massive MIMO system, is considered. Instead of the channel matrix, the channel model parameters are learned by neural networks to save the overhead and improve the accuracy of channel reconstruction. By viewing the channel as an image, we introduce You Only Look Once (YOLO), a powerful neural network for object detection, to enable a rapid estimation process of the model parameters, including the detection of angles and delays of the paths and the identification of visibility regions of the scatterers. The deep learning-based scheme avoids the complicated iterative process introduced by the algorithm-based parameter extraction methods. A low-complexity algorithm-based refiner further refines the YOLO estimates toward high accuracy. Given the efficiency of model-driven deep learning and the combination of neural network and algorithm, the proposed scheme can rapidly and accurately reconstruct the non-stationary downlink channel. Moreover, the proposed scheme is also applicable to widely concerned stationary systems and achieves comparable reconstruction accuracy as an algorithm-based method with greatly reduced time consumption.

Leveraging persona information of users in Neural Response Generators (NRG) to perform personalized conversations has been considered as an attractive and important topic in the research of conversational agents over the past few years. Despite of the promising progresses achieved by recent studies in this field, persona information tends to be incorporated into neural networks in the form of user embeddings, with the expectation that the persona can be involved via the End-to-End learning. This paper proposes to adopt the personality-related characteristics of human conversations into variational response generators, by designing a specific conditional variational autoencoder based deep model with two new regularization terms employed to the loss function, so as to guide the optimization towards the direction of generating both persona-aware and relevant responses. Besides, to reasonably evaluate the performances of various persona modeling approaches, this paper further presents three direct persona-oriented metrics from different perspectives. The experimental results have shown that our proposed methodology can notably improve the performance of persona-aware response generation, and the metrics are reasonable to evaluate the results.

Chinese meme-face is a special kind of internet subculture widely spread in Chinese Social Community Networks. It usually consists of a template image modified by some amusing details and a text caption. In this paper, we present MemeFaceGenerator, a Generative Adversarial Network with the attention module and template information as supplementary signals, to automatically generate meme-faces from text inputs. We also develop a web service as system demonstration of meme-face synthesis. MemeFaceGenerator has been shown to be capable of generating high-quality meme-faces from random text inputs.