Intelligent reflecting surface (IRS) can be densely deployed in complex environment to create cascaded line-of-sight (LoS) paths between multiple base stations (BSs) and users via tunable IRS reflections, thereby significantly enhancing the coverage performance of wireless networks. To achieve this goal, it is vital to optimize the deployed locations of BSs and IRSs in the wireless network, which is investigated in this paper. Specifically, we divide the coverage area of the network into multiple non-overlapping cells and decide whether to deploy a BS/IRS in each cell given a total number of BSs/IRSs available. We show that to ensure the network coverage/communication performance, i.e., each cell has a direct/cascaded LoS path with at least one BS, as well as such LoS paths have the average number of IRS reflections less than a given threshold, there is a fundamental trade-off with the deployment cost or the number of BSs/IRSs needed. To optimally characterize this trade-off, we formulate a joint BS and IRS deployment problem based on graph theory, which, however, is difficult to be optimally solved due to the combinatorial optimization involved. To circumvent this difficulty, we first consider a simplified problem with given BS deployment and propose the optimal as well as an efficient suboptimal IRS deployment solution to it, by applying the branch-and-bound method and iteratively removing IRSs from the candidate locations, respectively. Next, an efficient sequential update algorithm is proposed for solving the joint BS and IRS deployment problem. Numerical results are provided to show the efficacy of the proposed design approach and optimization algorithms for the joint BS and IRS deployment. The trade-off between the network coverage performance and the number of deployed BSs/IRSs with different cost ratios is also unveiled.
In this letter, we study the wireless point-to-point communication from a transmitter (Tx) to a receiver (Rx), which is jointly aided by an active intelligent reflecting surface (AIRS) and a passive IRS (PIRS). We consider two practical transmission schemes by deploying the two IRSs in different orders, namely, Tx$\rightarrow$PIRS$\rightarrow$AIRS$\rightarrow$Rx (TPAR) and Tx$\rightarrow$AIRS$\rightarrow$PIRS$\rightarrow$Rx (TAPR). Assuming line-of-sight channels, we derive the achievable rates for the two schemes by optimizing the placement of the AIRS with the location of the PIRS fixed. Our analysis shows that when the number of PIRS elements and/or the AIRS amplification power is small, the AIRS should be deployed closer to the Rx in both schemes, and TAPR outperforms TPAR with their respective optimized AIRS/PIRS placement. Simulation results validate our analysis and show the considerable performance gain achieved by the jointly optimized AIRS/PIRS deployment over the baseline double-PIRS system under the same power and IRS element budgets.
Many recent approaches to natural language tasks are built on the remarkable abilities of large language models. Large language models can perform in-context learning, where they learn a new task from a few task demonstrations, without any parameter updates. This work examines the implications of in-context learning for the creation of datasets for new natural language tasks. Departing from recent in-context learning methods, we formulate an annotation-efficient, two-step framework: selective annotation that chooses a pool of examples to annotate from unlabeled data in advance, followed by prompt retrieval that retrieves task examples from the annotated pool at test time. Based on this framework, we propose an unsupervised, graph-based selective annotation method, voke-k, to select diverse, representative examples to annotate. Extensive experiments on 10 datasets (covering classification, commonsense reasoning, dialogue, and text/code generation) demonstrate that our selective annotation method improves the task performance by a large margin. On average, vote-k achieves a 12.9%/11.4% relative gain under an annotation budget of 18/100, as compared to randomly selecting examples to annotate. Compared to state-of-the-art supervised finetuning approaches, it yields similar performance with 10-100x less annotation cost across 10 tasks. We further analyze the effectiveness of our framework in various scenarios: language models with varying sizes, alternative selective annotation methods, and cases where there is a test data domain shift. We hope that our studies will serve as a basis for data annotations as large language models are increasingly applied to new tasks. Our code is available at https://github.com/HKUNLP/icl-selective-annotation.
We present FOLIO, a human-annotated, open-domain, and logically complex and diverse dataset for reasoning in natural language (NL), equipped with first order logic (FOL) annotations. FOLIO consists of 1,435 examples (unique conclusions), each paired with one of 487 sets of premises which serve as rules to be used to deductively reason for the validity of each conclusion. The logical correctness of premises and conclusions is ensured by their parallel FOL annotations, which are automatically verified by our FOL inference engine. In addition to the main NL reasoning task, NL-FOL pairs in FOLIO automatically constitute a new NL-FOL translation dataset using FOL as the logical form. Our experiments on FOLIO systematically evaluate the FOL reasoning ability of supervised fine-tuning on medium-sized language models (BERT, RoBERTa) and few-shot prompting on large language models (GPT-NeoX, OPT, GPT-3, Codex). For NL-FOL translation, we experiment with GPT-3 and Codex. Our results show that one of the most capable Large Language Model (LLM) publicly available, GPT-3 davinci, achieves only slightly better than random results with few-shot prompting on a subset of FOLIO, and the model is especially bad at predicting the correct truth values for False and Unknown conclusions. Our dataset and code are available at https://github.com/Yale-LILY/FOLIO.
In this letter, we consider the fundamental problem of jointly designing the transmit beamformers and the phase- shifts of the intelligent reflecting surface (IRS) / reconfigurable intelligent surface (RIS) to minimize the transmit power, subject to quality-of-service constraints at individual users in an IRS- assisted multiuser multiple-input single-output downlink communication system. In particular, we propose a new successive convex approximation based second-order cone programming approach in which all the optimization variables are simultaneously updated in each iteration. Our proposed scheme achieves superior performance compared to state-of-the-art benchmark solutions. In addition, the complexity of the proposed scheme is $O(N_{\mathrm s}^{3.5})$, while that of state-of-the-art benchmark schemes is $O(N_{\mathrm s}^{7})$, where $N_{\mathrm s}$ denotes the number of reflecting elements at the IRS.
While exogenous variables have a major impact on performance improvement in time series analysis, inter-series correlation and time dependence among them are rarely considered in the present continuous methods. The dynamical systems of multivariate time series could be modelled with complex unknown partial differential equations (PDEs) which play a prominent role in many disciplines of science and engineering. In this paper, we propose a continuous-time model for arbitrary-step prediction to learn an unknown PDE system in multivariate time series whose governing equations are parameterised by self-attention and gated recurrent neural networks. The proposed model, \underline{E}xogenous-\underline{g}uided \underline{P}artial \underline{D}ifferential \underline{E}quation Network (EgPDE-Net), takes account of the relationships among the exogenous variables and their effects on the target series. Importantly, the model can be reduced into a regularised ordinary differential equation (ODE) problem with special designed regularisation guidance, which makes the PDE problem tractable to obtain numerical solutions and feasible to predict multiple future values of the target series at arbitrary time points. Extensive experiments demonstrate that our proposed model could achieve competitive accuracy over strong baselines: on average, it outperforms the best baseline by reducing $9.85\%$ on RMSE and $13.98\%$ on MAE for arbitrary-step prediction.
Comments, or natural language descriptions of source code, are standard practice among software developers. By communicating important aspects of the code such as functionality and usage, comments help with software project maintenance. However, when the code is modified without an accompanying correction to the comment, an inconsistency between the comment and code can arise, which opens up the possibility for developer confusion and bugs. In this paper, we propose two models based on BERT (Devlin et al., 2019) and Longformer (Beltagy et al., 2020) to detect such inconsistencies in a natural language inference (NLI) context. Through an evaluation on a previously established corpus of comment-method pairs both during and after code changes, we demonstrate that our models outperform multiple baselines and yield comparable results to the state-of-the-art models that exclude linguistic and lexical features. We further discuss ideas for future research in using pretrained language models for both inconsistency detection and automatic comment updating.
During the generation of invisible backdoor attack poisoned data, the feature space transformation operation tends to cause the loss of some poisoned features and weakens the mapping relationship between source images with triggers and target labels, resulting in the need for a higher poisoning rate to achieve the corresponding backdoor attack success rate. To solve the above problems, we propose the idea of feature repair for the first time and introduce the blind watermark technique to repair the poisoned features lost during the generation of poisoned data. Under the premise of ensuring consistent labeling, we propose a low-poisoning rate invisible backdoor attack based on feature repair, named FRIB. Benefiting from the above design concept, the new method enhances the mapping relationship between the source images with triggers and the target labels, and increases the degree of misleading DNNs, thus achieving a high backdoor attack success rate with a very low poisoning rate. Ultimately, the detailed experimental results show that the goal of achieving a high success rate of backdoor attacks with a very low poisoning rate is achieved on all MNIST, CIFAR10, GTSRB, and ImageNet datasets.
With the rapid development of satellite communication technologies, the space-based access network has been envisioned as a promising complementary part of the future 6G network. Aside from terrestrial base stations, satellite nodes, especially the low-earth-orbit (LEO) satellites, can also serve as base stations for Internet access, and constitute the LEO-satellite-based access network (LEO-SAN). LEO-SAN is expected to provide seamless massive access and extended coverage with high signal quality. However, its practical implementation still faces significant technical challenges, e.g., high mobility and limited budget for communication payloads of LEO satellite nodes. This paper aims at revealing the main technical issues that have not been fully addressed by the existing LEO-SAN designs, from three aspects namely random access, beam management and Doppler-resistant transmission technologies. More specifically, the critical issues of random access in LEO-SAN are discussed regarding low flexibility, long transmission delay, and inefficient handshakes. Then the beam management for LEO-SAN is investigated in complex propagation environments under the constraints of high mobility and limited payload budget. Furthermore, the influence of Doppler shifts on LEO-SAN is explored. Correspondingly, promising technologies to address these challenges are also discussed, respectively. Finally, the future research directions are envisioned.
Admittedly, Graph Convolution Network (GCN) has achieved excellent results on graph datasets such as social networks, citation networks, etc. However, softmax used as the decision layer in these frameworks is generally optimized with thousands of iterations via gradient descent. Furthermore, due to ignoring the inner distribution of the graph nodes, the decision layer might lead to an unsatisfactory performance in semi-supervised learning with less label support. To address the referred issues, we propose a novel graph deep model with a non-gradient decision layer for graph mining. Firstly, manifold learning is unified with label local-structure preservation to capture the topological information of the nodes. Moreover, owing to the non-gradient property, closed-form solutions is achieved to be employed as the decision layer for GCN. Particularly, a joint optimization method is designed for this graph model, which extremely accelerates the convergence of the model. Finally, extensive experiments show that the proposed model has achieved state-of-the-art performance compared to the current models.