Search

Next-generation channel coding has stringent demands on throughput, energy consumption, and error rate performance while maintaining key features of 5G New Radio (NR) standard codes such as rate compatibility, which is a significant challenge. Due to excellent capacity-achieving performance, spatially-coupled low-density parity-check (SC-LDPC) codes are considered a promising candidate for next-generation channel coding. In this paper, we propose an SC-LDPC code family called edge-spreading Raptor-like (ESRL) codes for 6G. Unlike other SC-LDPC codes that adopt the structure of existing rate-compatible LDPC block codes before coupling, ESRL codes maximize the possible locations of edge placement and focus on constructing an optimal coupled matrix. Moreover, a new graph representation called the unified graph is introduced. This graph offers a global perspective on ESRL codes and identifies the optimal edge reallocation to optimize the spreading strategy. We conduct comprehensive comparisons of ESRL codes and 5G-NR LDPC codes. Simulation results demonstrate that when all decoding parameters and complexity are the same, ESRL codes have obvious advantages in error rate performance and throughput compared to 5G-NR LDPC codes, making them a promising solution towards next-generation channel coding., Comment: 13 pages, 14 figures

Large Language Models (LLMs) have achieved great success in various reasoning tasks. In this work, we focus on the graph reasoning ability of LLMs. Although theoretical studies proved that LLMs are capable of handling graph reasoning tasks, empirical evaluations reveal numerous failures. To deepen our understanding on this discrepancy, we revisit the ability of LLMs on three fundamental graph tasks: graph description translation, graph connectivity, and the shortest-path problem. Our findings suggest that LLMs can fail to understand graph structures through text descriptions and exhibit varying performance for all these three fundamental tasks. Meanwhile, we perform a real-world investigation on knowledge graphs and make consistent observations with our findings. The codes and datasets are available.

Node-based successive cancellation list (SCL) decoding has received considerable attention in wireless communications for its significant reduction in decoding latency, particularly with 5G New Radio (NR) polar codes. However, the existing node-based SCL decoders are constrained by sequential processing, leading to complicated and data-dependent computational units that introduce unavoidable stalls, reducing hardware efficiency. In this paper, we present a frame-interleaving hardware architecture for a generalized node-based SCL decoder. By efficiently reusing otherwise idle computational units, two independent frames can be decoded simultaneously, resulting in a significant throughput gain. Based on this new architecture, we further exploit graph ensembles to diversify the decoding space, thus enhancing the error-correcting performance with a limited list size. Two dynamic strategies are proposed to eliminate the residual stalls in the decoding schedule, which eventually results in nearly 2x throughput compared to the state-of-the-art baseline node-based SCL decoder. To impart the decoder rate flexibility, we develop a novel online instruction generator to identify the generalized nodes and produce instructions on-the-fly. The corresponding 28nm FD-SOI ASIC SCL decoder with a list size of 8 has a core area of 1.28 mm2 and operates at 692 MHz. It is compatible with all 5G NR polar codes and achieves a throughput of 3.34 Gbps and an area efficiency of 2.62 Gbps/mm2 for uplink (1024, 512) codes, which is 1.41x and 1.69x better than the state-of-the-art node-based SCL decoders., Comment: 13 pages, 16 figures, accepted by IEEE Transactions on Circuits and Systems I: Regular Papers

Efficient fine-tuning plays a fundamental role in modern large models, with low-rank adaptation emerging as a particularly promising approach. However, the existing variants of LoRA are hampered by limited expressiveness, a tendency to overfit, and sensitivity to hyperparameter settings. This paper presents LoRA Slow Cascade Learning (LoRASC), an innovative technique designed to enhance LoRA's expressiveness and generalization capabilities while preserving its training efficiency. Our approach augments expressiveness through a cascaded learning strategy that enables a mixture-of-low-rank adaptation, thereby increasing the model's ability to capture complex patterns. Additionally, we introduce a slow-fast update mechanism and cascading noisy tuning to bolster generalization. The extensive experiments on various language and vision datasets, as well as robustness benchmarks, demonstrate that the proposed method not only significantly outperforms existing baselines, but also mitigates overfitting, enhances model stability, and improves OOD robustness. Code will be release in https://github.com/microsoft/LoRASC very soon.

Task planning in language agents is emerging as an important research topic alongside the development of large language models (LLMs). It aims to break down complex user requests in natural language into solvable sub-tasks, thereby fulfilling the original requests. In this context, the sub-tasks can be naturally viewed as a graph, where the nodes represent the sub-tasks, and the edges denote the dependencies among them. Consequently, task planning is a decision-making problem that involves selecting a connected path or subgraph within the corresponding graph and invoking it. In this paper, we explore graph learning-based methods for task planning, a direction that is orthogonal to the prevalent focus on prompt design. Our interest in graph learning stems from a theoretical discovery: the biases of attention and auto-regressive loss impede LLMs' ability to effectively navigate decision-making on graphs, which is adeptly addressed by graph neural networks (GNNs). This theoretical insight led us to integrate GNNs with LLMs to enhance overall performance. Extensive experiments demonstrate that GNN-based methods surpass existing solutions even without training, and minimal training can further enhance their performance. The performance gain increases with a larger task graph size., Comment: Accepted by NeurIPS 2024

In this paper, we present the findings of our Project ALPINE which stands for ``Autoregressive Learning for Planning In NEtworks." Project ALPINE initiates a theoretical investigation into the development of planning capabilities in Transformer-based language models through their autoregressive learning mechanisms, aiming to identify any potential limitations in their planning abilities. We abstract planning as a network path-finding task where the objective is to generate a valid path from a specified source node to a designated target node. In terms of expressiveness, we show that the Transformer is capable of executing path-finding by embedding the adjacency and reachability matrices within its weights. Our theoretical analysis of the gradient-based learning dynamic of the Transformer reveals that the Transformer is capable of learning both the adjacency matrix and a limited form of the reachability matrix. These theoretical insights are then validated through experiments, which demonstrate that the Transformer indeed learns the adjacency matrix and an incomplete reachability matrix, which aligns with the predictions made in our theoretical analysis. Additionally, when applying our methodology to a real-world planning benchmark, called Blocksworld, our observations remain consistent. Our theoretical and empirical analyses further unveil a potential limitation of Transformer in path-finding: it cannot identify reachability relationships through transitivity, and thus would fail when path concatenation is needed to generate a path. In summary, our findings shed new light on how the internal mechanisms of autoregressive learning enable planning in networks. This study may contribute to our understanding of the general planning capabilities in other related domains.

This paper focuses on the task of quality enhancement for compressed videos. Although deep network-based video restorers achieve impressive progress, most of the existing methods lack a structured design to optimally leverage the priors within compression codecs. Since the quality degradation of the video is primarily induced by the compression algorithm, a new paradigm is urgently needed for a more ``conscious'' process of quality enhancement. As a result, we propose the Compression-Realized Deep Structural Network (CRDS), introducing three inductive biases aligned with the three primary processes in the classic compression codec, merging the strengths of classical encoder architecture with deep network capabilities. Inspired by the residual extraction and domain transformation process in the codec, a pre-trained Latent Degradation Residual Auto-Encoder is proposed to transform video frames into a latent feature space, and the mutual neighborhood attention mechanism is integrated for precise motion estimation and residual extraction. Furthermore, drawing inspiration from the quantization noise distribution of the codec, CRDS proposes a novel Progressive Denoising framework with intermediate supervision that decomposes the quality enhancement into a series of simpler denoising sub-tasks. Experimental results on datasets like LDV 2.0 and MFQE 2.0 indicate our approach surpasses state-of-the-art models.

State space models and Mamba-based models have been increasingly applied across various domains, achieving state-of-the-art performance. This technical report introduces the first attempt to train a transferable Mamba model utilizing contrastive language-image pretraining (CLIP). We have trained Mamba models of varying sizes and undertaken comprehensive evaluations of these models on 26 zero-shot classification datasets and 16 out-of-distribution (OOD) datasets. Our findings reveal that a Mamba model with 67 million parameters is on par with a 307 million-parameter Vision Transformer (ViT) model in zero-shot classification tasks, highlighting the parameter efficiency of Mamba models. In tests of OOD generalization, Mamba-based models exhibit exceptional performance in conditions of OOD image contrast or when subjected to high-pass filtering. However, a Hessian analysis indicates that Mamba models feature a sharper and more non-convex landscape compared to ViT-based models, making them more challenging to train. The source code is available at https://github.com/raytrun/mamba-clip.

Adding additional control to pretrained diffusion models has become an increasingly popular research area, with extensive applications in computer vision, reinforcement learning, and AI for science. Recently, several studies have proposed training-free loss-based guidance by using off-the-shelf networks pretrained on clean images. This approach enables zero-shot conditional generation for universal control formats, which appears to offer a free lunch in diffusion guidance. In this paper, we aim to develop a deeper understanding of training-free guidance, as well as overcome its limitations. We offer a theoretical analysis that supports training-free guidance from the perspective of optimization, distinguishing it from classifier-based (or classifier-free) guidance. To elucidate their drawbacks, we theoretically demonstrate that training-free guidance is more susceptible to adversarial gradients and exhibits slower convergence rates compared to classifier guidance. We then introduce a collection of techniques designed to overcome the limitations, accompanied by theoretical rationale and empirical evidence. Our experiments in image and motion generation confirm the efficacy of these techniques.

Empowering embodied agents, such as robots, with Artificial Intelligence (AI) has become increasingly important in recent years. A major challenge is task open-endedness. In practice, robots often need to perform tasks with novel goals that are multifaceted, dynamic, lack a definitive "end-state", and were not encountered during training. To tackle this problem, this paper introduces \textit{Diffusion for Open-ended Goals} (DOG), a novel framework designed to enable embodied AI to plan and act flexibly and dynamically for open-ended task goals. DOG synergizes the generative prowess of diffusion models with state-of-the-art, training-free guidance techniques to adaptively perform online planning and control. Our evaluations demonstrate that DOG can handle various kinds of novel task goals not seen during training, in both maze navigation and robot control problems. Our work sheds light on enhancing embodied AI's adaptability and competency in tackling open-ended goals.

Label noise is a common challenge in large datasets, as it can significantly degrade the generalization ability of deep neural networks. Most existing studies focus on noisy labels in computer vision; however, graph models encompass both node features and graph topology as input, and become more susceptible to label noise through message-passing mechanisms. Recently, only a few works have been proposed to tackle the label noise on graphs. One significant limitation is that they operate under the assumption that the graph exhibits homophily and that the labels are distributed smoothly. However, real-world graphs can exhibit varying degrees of heterophily, or even be dominated by heterophily, which results in the inadequacy of the current methods. In this paper, we study graph label noise in the context of arbitrary heterophily, with the aim of rectifying noisy labels and assigning labels to previously unlabeled nodes. We begin by conducting two empirical analyses to explore the impact of graph homophily on graph label noise. Following observations, we propose a efficient algorithm, denoted as $R^{2}LP$. Specifically, $R^{2}LP$ is an iterative algorithm with three steps: (1) reconstruct the graph to recover the homophily property, (2) utilize label propagation to rectify the noisy labels, (3) select high-confidence labels to retain for the next iteration. By iterating these steps, we obtain a set of correct labels, ultimately achieving high accuracy in the node classification task. The theoretical analysis is also provided to demonstrate its remarkable denoising effect. Finally, we perform experiments on ten benchmark datasets with different levels of graph heterophily and various types of noise. In these experiments, we compare the performance of $R^{2}LP$ against ten typical baseline methods. Our results illustrate the superior performance of the proposed $R^{2}LP$., Comment: To appear in KDD2024

5G New Radio (NR) has stringent demands on both performance and complexity for the design of low-density parity-check (LDPC) decoding algorithms and corresponding VLSI implementations. Furthermore, decoders must fully support the wide range of all 5G NR blocklengths and code rates, which is a significant challenge. In this paper, we present a high-performance and low-complexity LDPC decoder, tailor-made to fulfill the 5G requirements. First, to close the gap between belief propagation (BP) decoding and its approximations in hardware, we propose an extension of adjusted min-sum decoding, called generalized adjusted min-sum (GA-MS) decoding. This decoding algorithm flexibly truncates the incoming messages at the check node level and carefully approximates the non-linear functions of BP decoding to balance the error-rate and hardware complexity. Numerical results demonstrate that the proposed fixed-point GAMS has only a minor gap of 0.1 dB compared to floating-point BP under various scenarios of 5G standard specifications. Secondly, we present a fully reconfigurable 5G NR LDPC decoder implementation based on GA-MS decoding. Given that memory occupies a substantial portion of the decoder area, we adopt multiple data compression and approximation techniques to reduce 42.2% of the memory overhead. The corresponding 28nm FD-SOI ASIC decoder has a core area of 1.823 mm2 and operates at 895 MHz. It is compatible with all 5G NR LDPC codes and achieves a peak throughput of 24.42 Gbps and a maximum area efficiency of 13.40 Gbps/mm2 at 4 decoding iterations., Comment: 14 pages, 15 figures, accepted by IEEE Transactions on Circuits and Systems I: Regular Paper

While language-guided image manipulation has made remarkable progress, the challenge of how to instruct the manipulation process faithfully reflecting human intentions persists. An accurate and comprehensive description of a manipulation task using natural language is laborious and sometimes even impossible, primarily due to the inherent uncertainty and ambiguity present in linguistic expressions. Is it feasible to accomplish image manipulation without resorting to external cross-modal language information? If this possibility exists, the inherent modality gap would be effortlessly eliminated. In this paper, we propose a novel manipulation methodology, dubbed ImageBrush, that learns visual instructions for more accurate image editing. Our key idea is to employ a pair of transformation images as visual instructions, which not only precisely captures human intention but also facilitates accessibility in real-world scenarios. Capturing visual instructions is particularly challenging because it involves extracting the underlying intentions solely from visual demonstrations and then applying this operation to a new image. To address this challenge, we formulate visual instruction learning as a diffusion-based inpainting problem, where the contextual information is fully exploited through an iterative process of generation. A visual prompting encoder is carefully devised to enhance the model's capacity in uncovering human intent behind the visual instructions. Extensive experiments show that our method generates engaging manipulation results conforming to the transformations entailed in demonstrations. Moreover, our model exhibits robust generalization capabilities on various downstream tasks such as pose transfer, image translation and video inpainting.

The evolution of wireless networks gravitates towards connected intelligence, a concept that envisions seamless interconnectivity among humans, objects, and intelligence in a hyper-connected cyber-physical world. Edge artificial intelligence (Edge AI) is a promising solution to achieve connected intelligence by delivering high-quality, low-latency, and privacy-preserving AI services at the network edge. This article presents a vision of autonomous edge AI systems that automatically organize, adapt, and optimize themselves to meet users' diverse requirements, leveraging the power of large language models (LLMs), i.e., Generative Pretrained Transformer (GPT). By exploiting the powerful abilities of GPT in language understanding, planning, and code generation, as well as incorporating classic wisdom such as task-oriented communication and edge federated learning, we present a versatile framework that efficiently coordinates edge AI models to cater to users' personal demands while automatically generating code to train new models in a privacy-preserving manner. Experimental results demonstrate the system's remarkable ability to accurately comprehend user demands, efficiently execute AI models with minimal cost, and effectively create high-performance AI models at edge servers., Comment: IEEE Communication Magazine

Diffusion-based Generative Models (DGMs) have achieved unparalleled performance in synthesizing high-quality visual content, opening up the opportunity to improve image super-resolution (SR) tasks. Recent solutions for these tasks often train architecture-specific DGMs from scratch, or require iterative fine-tuning and distillation on pre-trained DGMs, both of which take considerable time and hardware investments. More seriously, since the DGMs are established with a discrete pre-defined upsampling scale, they cannot well match the emerging requirements of arbitrary-scale super-resolution (ASSR), where a unified model adapts to arbitrary upsampling scales, instead of preparing a series of distinct models for each case. These limitations beg an intriguing question: can we identify the ASSR capability of existing pre-trained DGMs without the need for distillation or fine-tuning? In this paper, we take a step towards resolving this matter by proposing Diff-SR, a first ASSR attempt based solely on pre-trained DGMs, without additional training efforts. It is motivated by an exciting finding that a simple methodology, which first injects a specific amount of noise into the low-resolution images before invoking a DGM's backward diffusion process, outperforms current leading solutions. The key insight is determining a suitable amount of noise to inject, i.e., small amounts lead to poor low-level fidelity, while over-large amounts degrade the high-level signature. Through a finely-grained theoretical analysis, we propose the Perceptual Recoverable Field (PRF), a metric that achieves the optimal trade-off between these two factors. Extensive experiments verify the effectiveness, flexibility, and adaptability of Diff-SR, demonstrating superior performance to state-of-the-art solutions under diverse ASSR environments.

Prompt learning is one of the most effective and trending ways to adapt powerful vision-language foundation models like CLIP to downstream datasets by tuning learnable prompt vectors with very few samples. However, although prompt learning achieves excellent performance over in-domain data, it still faces the major challenge of generalizing to unseen classes and domains. Some existing prompt learning methods tackle this issue by adaptively generating different prompts for different tokens or domains but neglecting the ability of learned prompts to generalize to unseen domains. In this paper, we propose a novel prompt learning paradigm that directly generates \emph{domain invariant} prompt that can be generalized to unseen domains, called MetaPrompt. Specifically, a dual-modality prompt tuning network is proposed to generate prompts for input from both image and text modalities. With a novel asymmetric contrastive loss, the representation from the original pre-trained vision-language model acts as supervision to enhance the generalization ability of the learned prompt. More importantly, we propose a meta-learning-based prompt tuning algorithm that explicitly constrains the task-specific prompt tuned for one domain or class to also achieve good performance in another domain or class. Extensive experiments on 11 datasets for base-to-new generalization and 4 datasets for domain generalization demonstrate that our method consistently and significantly outperforms existing methods., Comment: 12 pages, 6 figures, 5 tables

Terahertz ultra-massive MIMO (THz UM-MIMO) is envisioned as one of the key enablers of 6G wireless networks, for which channel estimation is highly challenging. Traditional analytical estimation methods are no longer effective, as the enlarged array aperture and the small wavelength result in a mixture of far-field and near-field paths, constituting a hybrid-field channel. Deep learning (DL)-based methods, despite the competitive performance, generally lack theoretical guarantees and scale poorly with the size of the array. In this paper, we propose a general DL framework for THz UM-MIMO channel estimation, which leverages existing iterative channel estimators and is with provable guarantees. Each iteration is implemented by a fixed point network (FPN), consisting of a closed-form linear estimator and a DL-based non-linear estimator. The proposed method perfectly matches the THz UM-MIMO channel estimation due to several unique advantages. First, the complexity is low and adaptive. It enjoys provable linear convergence with a low per-iteration cost and monotonically increasing accuracy, which enables an adaptive accuracy-complexity tradeoff. Second, it is robust to practical distribution shifts and can directly generalize to a variety of heavily out-of-distribution scenarios with almost no performance loss, which is suitable for the complicated THz channel conditions. For practical usage, the proposed framework is further extended to wideband THz UM-MIMO systems with beam squint effect. Theoretical analysis and extensive simulation results are provided to illustrate the advantages over the state-of-the-art methods in estimation accuracy, convergence rate, complexity, and robustness., Comment: 15 pages, 11 figures, 5 tables, accepted by IEEE Journal of Selected Topics in Signal Processing (JSTSP)

Owing to its high parallelism, belief propagation (BP) decoding is highly amenable to high-throughput implementations and thus represents a promising solution for meeting the ultra-high peak data rate of future communication systems. However, for polar codes, the error-correcting performance of BP decoding is far inferior to that of the widely used CRC-aided successive cancellation list (SCL) decoding algorithm. To close the performance gap to SCL, BP list (BPL) decoding expands the exploration of candidate codewords through multiple permuted factor graphs (PFGs). From an implementation perspective, designing a unified and flexible hardware architecture for BPL decoding that supports various PFGs and code configurations presents a big challenge. In this paper, we propose the first hardware implementation of a BPL decoder for polar codes and overcome the implementation challenge by applying a hardware-friendly algorithm that generates flexible permutations on-the-fly. First, we derive the graph selection gain and provide a sequential generation (SG) algorithm to obtain a near-optimal PFG set. We further prove that any permutation can be decomposed into a combination of multiple fixed routings, and we design a low-complexity permutation network to satisfy the decoding schedule. Our BPL decoder not only has a low decoding latency by executing the decoding and permutation generation in parallel, but also supports an arbitrary list size without any area overhead. Experimental results show that, for length-1024 polar codes with a code rate of one-half, our BPL decoder with 32 PFGs has a similar error-correcting performance to SCL with a list size of 4 and achieves a throughput of 25.63 Gbps and an area efficiency of 29.46 Gbps/mm$^{2}$ at SNR=4.0dB, which is 1.82$\times$ and 4.33$\times$ faster than the state-of-the-art BP flip and SCL decoders,~respectively

Compared to the linear MIMO detectors, the Belief Propagation (BP) detector has shown greater capabilities in achieving near optimal performance and better nature to iteratively cooperate with channel decoders. Aiming at real applications, recent works mainly fall into the category of reducing the complexity by simplified calculations, at the expense of performance sacrifice. However, the complexity is still unsatisfactory with exponentially increasing complexity or required exponentiation operations. Furthermore, due to the inherent loopy structure, the existing BP detectors persistently encounter error floor in high signal-to-noise ratio (SNR) region, which becomes even worse with calculation approximation. This work aims at a revised BP detector, named {Belief-selective Propagation (BsP)} detector by selectively utilizing the \emph{trusted} incoming messages with sufficiently large \textit{a priori} probabilities for updates. Two proposed strategies: symbol-based truncation (ST) and edge-based simplification (ES) squeeze the complexity (orders lower than the Original-BP), while greatly relieving the error floor issue over a wide range of antenna and modulation combinations. For the $16$-QAM $8 \times 4$ MIMO system, the $\mathcal{B}(1,1)$ {BsP} detector achieves more than $4$\,dB performance gain (@$\text{BER}=10^{-4}$) with roughly $4$ orders lower complexity than the Original-BP detector. Trade-off between performance and complexity towards different application requirement can be conveniently obtained by configuring the ST and ES parameters.

Human visual perception can easily generalize to out-of-distributed visual data, which is far beyond the capability of modern machine learning models. Domain generalization (DG) aims to close this gap, with existing DG methods mainly focusing on the loss function design. In this paper, we propose to explore an orthogonal direction, i.e., the design of the backbone architecture. It is motivated by an empirical finding that transformer-based models trained with empirical risk minimization (ERM) outperform CNN-based models employing state-of-the-art (SOTA) DG algorithms on multiple DG datasets. We develop a formal framework to characterize a network's robustness to distribution shifts by studying its architecture's alignment with the correlations in the dataset. This analysis guides us to propose a novel DG model built upon vision transformers, namely Generalizable Mixture-of-Experts (GMoE). Extensive experiments on DomainBed demonstrate that GMoE trained with ERM outperforms SOTA DG baselines by a large margin. Moreover, GMoE is complementary to existing DG methods and its performance is substantially improved when trained with DG algorithms., Comment: ICLR 2023 (accepted as Oral presentation)

Due to the low-latency and high-reliability requirements of 5G, low-complexity node-based successive cancellation list (SCL) decoding has received considerable attention for use in 5G communications systems. By identifying special constituent codes in the decoding tree and immediately decoding these, node-based SCL decoding provides a significant reduction in decoding latency compared to conventional SCL decoding. However, while there exists many types of nodes, the current node-based SCL decoders are limited by the lack of a more generalized node that can efficiently decode a larger number of different constituent codes to further reduce the decoding time. In this paper, we extend a recent generalized node, the sequence repetition (SR) node to SCL decoding and we describe the first implementation of an SR-List decoder. By merging certain SR-List decoding operations and applying various optimizations for 5G New Radio (NR) polar codes, our optimized SR-List decoding algorithm increases the throughput by almost ${2\times}$ compared to a similar state-of-the-art node-based SCL decoder. We also present our hardware implementation of the optimized SR-List decoding algorithm which supports all 5G NR polar codes. Synthesis results show that our SR-List decoder can achieve a $2.94 \, \mathrm{Gbps}$ throughput and $6.70\, \mathrm{Gbps} / \mathrm{mm}^2$ area efficiency for ${L=8}$.

Terahertz ultra-massive multiple-input multiple-output (THz UM-MIMO) is envisioned as one of the key enablers of 6G wireless systems. Due to the joint effect of its array aperture and small wavelength, the near-field region of THz UM-MIMO is greatly enlarged. The high-dimensional channel of such systems thus consists of a stochastic mixture of far and near fields, which renders channel estimation extremely challenging. Previous works based on uni-field assumptions cannot capture the hybrid far- and near-field features, thus suffering significant performance loss. This motivates us to consider hybrid-field channel estimation. We draw inspirations from fixed point theory to develop an efficient deep learning based channel estimator with adaptive complexity and linear convergence guarantee. Built upon classic orthogonal approximate message passing, we transform each iteration into a contractive mapping, comprising a closed-form linear estimator and a neural network based non-linear estimator. A major algorithmic innovation involves applying fixed point iteration to compute the channel estimate while modeling neural networks with arbitrary depth and adapting to the hybrid-field channel conditions. Simulation results verify our theoretical analysis and show significant performance gains over state-of-the-art approaches in the estimation accuracy and convergence rate., Comment: 6 pages, 3 figures, accepted by IEEE Globecom 2022. Source code is publicly available at https://github.com/wyuaq/FPN-OAMP-THz-Channel-Estimation

Deep learning-based approaches have been developed to solve challenging problems in wireless communications, leading to promising results. Early attempts adopted neural network architectures inherited from applications such as computer vision. They often yield poor performance in large scale networks (i.e., poor scalability) and unseen network settings (i.e., poor generalization). To resolve these issues, graph neural networks (GNNs) have been recently adopted, as they can effectively exploit the domain knowledge, i.e., the graph topology in wireless communications problems. GNN-based methods can achieve near-optimal performance in large-scale networks and generalize well under different system settings, but the theoretical underpinnings and design guidelines remain elusive, which may hinder their practical implementations. This paper endeavors to fill both the theoretical and practical gaps. For theoretical guarantees, we prove that GNNs achieve near-optimal performance in wireless networks with much fewer training samples than traditional neural architectures. Specifically, to solve an optimization problem on an $n$-node graph (where the nodes may represent users, base stations, or antennas), GNNs' generalization error and required number of training samples are $\mathcal{O}(n)$ and $\mathcal{O}(n^2)$ times lower than the unstructured multi-layer perceptrons. For design guidelines, we propose a unified framework that is applicable to general design problems in wireless networks, which includes graph modeling, neural architecture design, and theory-guided performance enhancement. Extensive simulations, which cover a variety of important problems and network settings, verify our theory and the effectiveness of the proposed design framework.

This technical note describes the recent updates of Graphormer, including architecture design modifications, and the adaption to 3D molecular dynamics simulation. With these simple modifications, Graphormer could attain better results on large-scale molecular modeling datasets than the vanilla one, and the performance gain could be consistently obtained on 2D and 3D molecular graph modeling tasks. In addition, we show that with a global receptive field and an adaptive aggregation strategy, Graphormer is more powerful than classic message-passing-based GNNs. Empirically, Graphormer could achieve much less MAE than the originally reported results on the PCQM4M quantum chemistry dataset used in KDD Cup 2021. In the meanwhile, it greatly outperforms the competitors in the recent Open Catalyst Challenge, which is a competition track on NeurIPS 2021 workshop, and aims to model the catalyst-adsorbate reaction system with advanced AI models. All codes could be found at https://github.com/Microsoft/Graphormer.

This technical note describes the recent updates of Graphormer, including architecture design modifications, and the adaption to 3D molecular dynamics simulation. The "Graphormer-V2" could attain better results on large-scale molecular modeling datasets than the vanilla one, and the performance gain could be consistently obtained on downstream tasks. In addition, we show that with a global receptive field and an adaptive aggregation strategy, Graphormer is more powerful than classic message-passing-based GNNs. Graphormer-V2 achieves much less MAE than the vanilla Graphormer on the PCQM4M quantum chemistry dataset used in KDD Cup 2021, where the latter one won the first place in this competition. In the meanwhile, Graphormer-V2 greatly outperforms the competitors in the recent Open Catalyst Challenge, which is a competition track on NeurIPS 2021 workshop, and aims to model the catalyst-adsorbate reaction system with advanced AI models. All models could be found at \url{https://github.com/Microsoft/Graphormer}., Comment: Wrong dual-submission (arXiv:2203.04810) with negligently

To fully support vertical industries, 5G and its corresponding channel coding are expected to meet requirements of different applications. However, for applications of 5G and beyond 5G (B5G) such as URLLC, the transmission latency is required to be much shorter than that in eMBB. Therefore, the resulting channel code length reduces drastically. In this case, the traditional 1-D channel coding suffers a lot from the performance degradation and fails to deliver strong reliability with very low latency. To remove this bottleneck, new channel coding scheme beyond the existing 1-D one is in urgent need. By making full use of the spacial freedom of massive MIMO systems, this paper devotes itself in proposing a spatiotemporal 2-D channel coding for very low latency reliable transmission. For a very short time-domain code length $N^{\text{time}}=16$, $64 \times 128$ MIMO system employing the proposed spatiotemporal 2-D coding scheme successfully shows more than $3$\,dB performance gain at $\text{FER}=10^{-3}$, compared to the 1-D time-domain channel coding. It is noted that the proposed coding scheme is suitable for different channel codes and enjoys high flexibility to adapt to difference scenarios. By appropriately selecting the code rate, code length, and the number of codewords in the time and space domains, the proposed coding scheme can achieve a good trade-off between the transmission latency and reliability.

Continuous-depth neural networks, such as the Neural Ordinary Differential Equations (ODEs), have aroused a great deal of interest from the communities of machine learning and data science in recent years, which bridge the connection between deep neural networks and dynamical systems. In this article, we introduce a new sort of continuous-depth neural network, called the Neural Piecewise-Constant Delay Differential Equations (PCDDEs). Here, unlike the recently proposed framework of the Neural Delay Differential Equations (DDEs), we transform the single delay into the piecewise-constant delay(s). The Neural PCDDEs with such a transformation, on one hand, inherit the strength of universal approximating capability in Neural DDEs. On the other hand, the Neural PCDDEs, leveraging the contributions of the information from the multiple previous time steps, further promote the modeling capability without augmenting the network dimension. With such a promotion, we show that the Neural PCDDEs do outperform the several existing continuous-depth neural frameworks on the one-dimensional piecewise-constant delay population dynamics and real-world datasets, including MNIST, CIFAR10, and SVHN., Comment: 9pages, 5 figures, accepted in AAAI 2022

In this paper, we consider a different data format for images: vector graphics. In contrast to raster graphics which are widely used in image recognition, vector graphics can be scaled up or down into any resolution without aliasing or information loss, due to the analytic representation of the primitives in the document. Furthermore, vector graphics are able to give extra structural information on how low-level elements group together to form high level shapes or structures. These merits of graphic vectors have not been fully leveraged in existing methods. To explore this data format, we target on the fundamental recognition tasks: object localization and classification. We propose an efficient CNN-free pipeline that does not render the graphic into pixels (i.e. rasterization), and takes textual document of the vector graphics as input, called YOLaT (You Only Look at Text). YOLaT builds multi-graphs to model the structural and spatial information in vector graphics, and a dual-stream graph neural network is proposed to detect objects from the graph. Our experiments show that by directly operating on vector graphics, YOLaT out-performs raster-graphic based object detection baselines in terms of both average precision and efficiency., Comment: Accepted by NeurIPS2021

In recent years, there has been a surge in applying deep learning to various challenging design problems in communication networks. The early attempts adopt neural architectures inherited from applications such as computer vision, which suffer from poor generalization, scalability, and lack of interpretability. To tackle these issues, domain knowledge has been integrated into the neural architecture design, which achieves near-optimal performance in large-scale networks and generalizes well under different system settings. This paper endeavors to theoretically validate the importance and effects of neural architectures when applying deep learning to design communication networks. We prove that by exploiting permutation invariance, a common property in communication networks, graph neural networks (GNNs) converge faster and generalize better than fully connected multi-layer perceptrons (MLPs), especially when the number of nodes (e.g., users, base stations, or antennas) is large. Specifically, we prove that under common assumptions, for a communication network with $n$ nodes, GNNs converge $O(n \log n)$ times faster and their generalization error is $O(n)$ times lower, compared with MLPs.

The conventional design of wireless communication systems typically relies on established mathematical models that capture the characteristics of different communication modules. Unfortunately, such design cannot be easily and directly applied to future wireless networks, which will be characterized by large-scale ultra-dense networks whose design complexity scales exponentially with the network size. Furthermore, such networks will vary dynamically in a significant way, which makes it intractable to develop comprehensive analytical models. Recently, deep learning-based approaches have emerged as potential alternatives for designing complex and dynamic wireless systems. However, existing learning-based methods have limited capabilities to scale with the problem size and to generalize with varying network settings. In this paper, we propose a scalable and generalizable neural calibration framework for future wireless system design, where a neural network is adopted to calibrate the input of conventional model-based algorithms. Specifically, the backbone of a traditional time-efficient algorithm is integrated with deep neural networks to achieve a high computational efficiency, while enjoying enhanced performance. The permutation equivariance property, carried out by the topological structure of wireless systems, is furthermore utilized to develop a generalizable neural network architecture. The proposed neural calibration framework is applied to solve challenging resource management problems in massive multiple-input multiple-output (MIMO) systems. Simulation results will show that the proposed neural calibration approach enjoys significantly improved scalability and generalization compared with the existing learning-based methods., Comment: submitted to IEEE Transactions on Wireless Communications. arXiv admin note: text overlap with arXiv:2108.01529

Graph convolutional networks (GCNs) have recently enabled a popular class of algorithms for collaborative filtering (CF). Nevertheless, the theoretical underpinnings of their empirical successes remain elusive. In this paper, we endeavor to obtain a better understanding of GCN-based CF methods via the lens of graph signal processing. By identifying the critical role of smoothness, a key concept in graph signal processing, we develop a unified graph convolution-based framework for CF. We prove that many existing CF methods are special cases of this framework, including the neighborhood-based methods, low-rank matrix factorization, linear auto-encoders, and LightGCN, corresponding to different low-pass filters. Based on our framework, we then present a simple and computationally efficient CF baseline, which we shall refer to as Graph Filter based Collaborative Filtering (GF-CF). Given an implicit feedback matrix, GF-CF can be obtained in a closed form instead of expensive training with back-propagation. Experiments will show that GF-CF achieves competitive or better performance against deep learning-based methods on three well-known datasets, notably with a $70\%$ performance gain over LightGCN on the Amazon-book dataset.

Channel estimation and beamforming play critical roles in frequency-division duplexing (FDD) massive multiple-input multiple-output (MIMO) systems. However, these two modules have been treated as two stand-alone components, which makes it difficult to achieve a global system optimality. In this paper, we propose a deep learning-based approach that directly optimizes the beamformers at the base station according to the received uplink pilots, thereby, bypassing the explicit channel estimation. Different from the existing fully data-driven approach where all the modules are replaced by deep neural networks (DNNs), a neural calibration method is proposed to improve the scalability of the end-to-end design. In particular, the backbone of conventional time-efficient algorithms, i.e., the least-squares (LS) channel estimator and the zero-forcing (ZF) beamformer, is preserved and DNNs are leveraged to calibrate their inputs for better performance. The permutation equivariance property of the formulated resource allocation problem is then identified to design a low-complexity neural network architecture. Simulation results will show the superiority of the proposed neural calibration method over benchmark schemes in terms of both the spectral efficiency and scalability in large-scale wireless networks., Comment: submitted to IEEE Globecom 2021

Invariant risk minimization (IRM) has recently emerged as a promising alternative for domain generalization. Nevertheless, the loss function is difficult to optimize for nonlinear classifiers and the original optimization objective could fail when pseudo-invariant features and geometric skews exist. Inspired by IRM, in this paper we propose a novel formulation for domain generalization, dubbed invariant information bottleneck (IIB). IIB aims at minimizing invariant risks for nonlinear classifiers and simultaneously mitigating the impact of pseudo-invariant features and geometric skews. Specifically, we first present a novel formulation for invariant causal prediction via mutual information. Then we adopt the variational formulation of the mutual information to develop a tractable loss function for nonlinear classifiers. To overcome the failure modes of IRM, we propose to minimize the mutual information between the inputs and the corresponding representations. IIB significantly outperforms IRM on synthetic datasets, where the pseudo-invariant features and geometric skews occur, showing the effectiveness of proposed formulation in overcoming failure modes of IRM. Furthermore, experiments on DomainBed show that IIB outperforms $13$ baselines by $0.9\%$ on average across $7$ real datasets., Comment: AAAI 2022

Resource management plays a pivotal role in wireless networks, which, unfortunately, leads to challenging NP-hard problems. Artificial Intelligence (AI), especially deep learning techniques, has recently emerged as a disruptive technology to solve such challenging problems in a real-time manner. However, although promising results have been reported, practical design guidelines and performance guarantees of AI-based approaches are still missing. In this paper, we endeavor to address two fundamental questions: 1) What are the main advantages of AI-based methods compared with classical techniques; and 2) Which neural network should we choose for a given resource management task. For the first question, four advantages are identified and discussed. For the second question, \emph{optimality gap}, i.e., the gap to the optimal performance, is proposed as a measure for selecting model architectures, as well as, for enabling a theoretical comparison between different AI-based approaches. Specifically, for $K$-user interference management problem, we theoretically show that graph neural networks (GNNs) are superior to multi-layer perceptrons (MLPs), and the performance gap between these two methods grows with $\sqrt{K}$., Comment: Meditcom 2021

In this paper, we investigate the decentralized statistical inference problem, where a network of agents cooperatively recover a (structured) vector from private noisy samples without centralized coordination. Existing optimization-based algorithms suffer from issues of model mismatch and poor convergence speed, and thus their performance would be degraded, provided that the number of communication rounds is limited. This motivates us to propose a learning-based framework, which unrolls well-noted decentralized optimization algorithms (e.g., Prox-DGD and PG-EXTRA) into graph neural networks (GNNs). By minimizing the recovery error via end-to-end training, this learning-based framework resolves the model mismatch issue. Our convergence analysis (with PG-EXTRA as the base algorithm) reveals that the learned model parameters may accelerate the convergence and reduce the recovery error to a large extent. The simulation results demonstrate that the proposed GNN-based learning methods prominently outperform several state-of-the-art optimization-based algorithms in convergence speed and recovery error.