Existing learning-based stereo image codec adopt sophisticated transformation with simple entropy models derived from single image codecs to encode latent representations. However, those entropy models struggle to effectively capture the spatial-disparity characteristics inherent in stereo images, which leads to suboptimal rate-distortion results. In this paper, we propose a stereo image compression framework, named CAMSIC. CAMSIC independently transforms each image to latent representation and employs a powerful decoder-free Transformer entropy model to capture both spatial and disparity dependencies, by introducing a novel content-aware masked image modeling (MIM) technique. Our content-aware MIM facilitates efficient bidirectional interaction between prior information and estimated tokens, which naturally obviates the need for an extra Transformer decoder. Experiments show that our stereo image codec achieves state-of-the-art rate-distortion performance on two stereo image datasets Cityscapes and InStereo2K with fast encoding and decoding speed.
Reconfigurable intelligent surface (RIS) is a novel meta-material which can form a smart radio environment by dynamically altering reflection directions of the impinging electromagnetic waves. In the prior literature, the inter-RIS links which also contribute to the performance of the whole system are usually neglected when multiple RISs are deployed. In this paper we investigate a general double-RIS assisted multiple-input multiple-output (MIMO) wireless communication system under spatially correlated non line-of-sight propagation channels, where the cooperation of the double RISs is also considered. The design objective is to maximize the achievable ergodic rate based on full statistical channel state information (CSI). Specifically, we firstly present a closed-form asymptotic expression for the achievable ergodic rate by utilizing replica method from statistical physics. Then a full statistical CSI-enabled optimal design is proposed which avoids high pilot training overhead compared to instantaneous CSI-enabled design. To further reduce the signal processing overhead and lower the complexity for practical realization, a common-phase scheme is proposed to design the double RISs. Simulation results show that the derived asymptotic ergodic rate is quite accurate even for small-sized antenna arrays. And the proposed optimization algorithm can achieve substantial gain at the expense of a low overhead and complexity. Furthermore, the cooperative double-RIS assisted MIMO framework is proven to achieve superior ergodic rate performance and high communication reliability under harsh propagation environment.
Implicit neural representations (INRs) have emerged as a promising approach for video storage and processing, showing remarkable versatility across various video tasks. However, existing methods often fail to fully leverage their representation capabilities, primarily due to inadequate alignment of intermediate features during target frame decoding. This paper introduces a universal boosting framework for current implicit video representation approaches. Specifically, we utilize a conditional decoder with a temporal-aware affine transform module, which uses the frame index as a prior condition to effectively align intermediate features with target frames. Besides, we introduce a sinusoidal NeRV-like block to generate diverse intermediate features and achieve a more balanced parameter distribution, thereby enhancing the model's capacity. With a high-frequency information-preserving reconstruction loss, our approach successfully boosts multiple baseline INRs in the reconstruction quality and convergence speed for video regression, and exhibits superior inpainting and interpolation results. Further, we integrate a consistent entropy minimization technique and develop video codecs based on these boosted INRs. Experiments on the UVG dataset confirm that our enhanced codecs significantly outperform baseline INRs and offer competitive rate-distortion performance compared to traditional and learning-based codecs.
Evolutionary algorithms, such as Differential Evolution, excel in solving real-parameter optimization challenges. However, the effectiveness of a single algorithm varies across different problem instances, necessitating considerable efforts in algorithm selection or configuration. This paper aims to address the limitation by leveraging the complementary strengths of a group of algorithms and dynamically scheduling them throughout the optimization progress for specific problems. We propose a deep reinforcement learning-based dynamic algorithm selection framework to accomplish this task. Our approach models the dynamic algorithm selection a Markov Decision Process, training an agent in a policy gradient manner to select the most suitable algorithm according to the features observed during the optimization process. To empower the agent with the necessary information, our framework incorporates a thoughtful design of landscape and algorithmic features. Meanwhile, we employ a sophisticated deep neural network model to infer the optimal action, ensuring informed algorithm selections. Additionally, an algorithm context restoration mechanism is embedded to facilitate smooth switching among different algorithms. These mechanisms together enable our framework to seamlessly select and switch algorithms in a dynamic online fashion. Notably, the proposed framework is simple and generic, offering potential improvements across a broad spectrum of evolutionary algorithms. As a proof-of-principle study, we apply this framework to a group of Differential Evolution algorithms. The experimental results showcase the remarkable effectiveness of the proposed framework, not only enhancing the overall optimization performance but also demonstrating favorable generalization ability across different problem classes.
Multi-talker automatic speech recognition plays a crucial role in scenarios involving multi-party interactions, such as meetings and conversations. Due to its inherent complexity, this task has been receiving increasing attention. Notably, the serialized output training (SOT) stands out among various approaches because of its simplistic architecture and exceptional performance. However, the frequent speaker changes in token-level SOT (t-SOT) present challenges for the autoregressive decoder in effectively utilizing context to predict output sequences. To address this issue, we introduce a masked t-SOT label, which serves as the cornerstone of an auxiliary training loss. Additionally, we utilize a speaker similarity matrix to refine the self-attention mechanism of the decoder. This strategic adjustment enhances contextual relationships within the same speaker's tokens while minimizing interactions between different speakers' tokens. We denote our method as speaker-aware SOT (SA-SOT). Experiments on the Librispeech datasets demonstrate that our SA-SOT obtains a relative cpWER reduction ranging from 12.75% to 22.03% on the multi-talker test sets. Furthermore, with more extensive training, our method achieves an impressive cpWER of 3.41%, establishing a new state-of-the-art result on the LibrispeechMix dataset.
Grant-free random access (RA) has been recognized as a promising solution to support massive connectivity due to the removal of the uplink grant request procedures. While most endeavours assume perfect synchronization among users and the base station, this paper investigates asynchronous grant-free massive RA, and develop efficient algorithms for joint user activity detection, synchronization delay detection, and channel estimation. Considering the sparsity on user activity, we formulate a sparse signal recovery problem and propose to utilize the framework of orthogonal approximate message passing (OAMP) to deal with the non-independent and identically distributed (i.i.d.) Gaussian pilot matrices caused by the synchronization delays. In particular, an OAMP-based algorithm is developed to fully harness the common sparsity among received pilot signals from multiple base station antennas. To reduce the computational complexity, we further propose a free probability AMP (FPAMP)-based algorithm, which exploits the rectangular free cumulants to make the cost-effective AMP framework compatible to general pilot matrices. Simulation results demonstrate that the two proposed algorithms outperform various baselines, and the FPAMP-based algorithm reduces 40% of the computations while maintaining comparable detection/estimation accuracy with the OAMP-based algorithm.
The ability of Large Language Models (LLMs) to process and generate coherent text is markedly weakened when the number of input tokens exceeds their pretraining length. Given the expensive overhead of finetuning large-scale models with longer sequences, we propose Dual Chunk Attention (DCA), which enables Llama2 70B to support context windows of more than 100k tokens without continual training. By decomposing the attention computation for long sequences into chunk-based modules, DCA manages to effectively capture the relative positional information of tokens within the same chunk (Intra-Chunk) and across distinct chunks (Inter-Chunk), as well as integrates seamlessly with Flash Attention. In addition to its impressive extrapolation capability, DCA achieves performance on practical long-context tasks that is comparable to or even better than that of finetuned models. When compared with proprietary models, our training-free 70B model attains 94% of the performance of gpt-3.5-16k, indicating it is a viable open-source alternative. All code and data used in this work are released at \url{https://github.com/HKUNLP/ChunkLlama}.
Federated learning (FL) is an emerging distributed training paradigm that aims to learn a common global model without exchanging or transferring the data that are stored locally at different clients. The Federated Averaging (FedAvg)-based algorithms have gained substantial popularity in FL to reduce the communication overhead, where each client conducts multiple localized iterations before communicating with a central server. In this paper, we focus on FL where the clients have diverse computation and/or communication capabilities. Under this circumstance, FedAvg can be less efficient since it requires all clients that participate in the global aggregation in a round to initiate iterations from the latest global model, and thus the synchronization among fast clients and straggler clients can severely slow down the overall training process. To address this issue, we propose an efficient asynchronous federated learning (AFL) framework called Delayed Federated Averaging (DeFedAvg). In DeFedAvg, the clients are allowed to perform local training with different stale global models at their own paces. Theoretical analyses demonstrate that DeFedAvg achieves asymptotic convergence rates that are on par with the results of FedAvg for solving nonconvex problems. More importantly, DeFedAvg is the first AFL algorithm that provably achieves the desirable linear speedup property, which indicates its high scalability. Additionally, we carry out extensive numerical experiments using real datasets to validate the efficiency and scalability of our approach when training deep neural networks.
This paper investigates resource allocation to provide heterogeneous users with customized virtual reality (VR) services in a mobile edge computing (MEC) system. We first introduce a quality of experience (QoE) metric to measure user experience, which considers the MEC system's latency, user attention levels, and preferred resolutions. Then, a QoE maximization problem is formulated for resource allocation to ensure the highest possible user experience,which is cast as a reinforcement learning problem, aiming to learn a generalized policy applicable across diverse user environments for all MEC servers. To learn the generalized policy, we propose a framework that employs federated learning (FL) and prompt-based sequence modeling to pre-train a common decision model across MEC servers, which is named FedPromptDT. Using FL solves the problem of insufficient local MEC data while protecting user privacy during offline training. The design of prompts integrating user-environment cues and user-preferred allocation improves the model's adaptability to various user environments during online execution.