Search

Proliferation of the urban Internet-of-Things (IoTs) for smart cities has fuelled massive amounts of data over wireless cellular networks. Random access (RA) system of wireless cellular networks, e.g., 5G New Radio (NR), based on S-ALOHA system should cope with ever-growing IoT traffic. This work proposes S-ALOHA system with time offsets (TOs), where one slot consists of K TOs and one packet transmission time. The length of the overall TOs is a fraction of a packet transmission time. In the system users (re)transmit to the boundary of a TO randomly selected. This enables the base station (BS) to inform the users of who transmits the first and the last packets in the slot with collision so that the two users can retransmit successfully in the following two slots respectively. Our throughput analysis compared to simulations shows that adopting even with three and four TOs surpasses the throughput limit of S-ALOHA system without TOs. Additionally, we propose two Bayesian-optimized backoff algorithms for S-ALOHA system with TOs, with which users can apply throughput-optimal (re)transmission probability or uniform backoff window even in unsaturated traffic scenarios. Numerical results demonstrate that the proposed backoff algorithms can achieve the throughput close to an ideal system and drastically reduce the access delay compared to S-ALOHA system., Comment: 17 pages, 26 figures. This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

Few-shot learning allows machines to classify novel classes using only a few labeled samples. Recently, few-shot segmentation aiming at semantic segmentation on low sample data has also seen great interest. In this paper, we propose a learnable module that can be placed on top of existing segmentation networks for performing few-shot segmentation. This module, called the task-adaptive feature transformer (TAFT), linearly transforms task-specific high-level features to a set of task agnostic features well-suited to conducting few-shot segmentation. The task-conditioned feature transformation allows an effective utilization of the semantic information in novel classes to generate tight segmentation masks. We also propose a semantic enrichment (SE) module that utilizes a pixel-wise attention module for high-level feature and an auxiliary loss from an auxiliary segmentation network conducting the semantic segmentation for all training classes. Experiments on PASCAL-$5^i$ and COCO-$20^i$ datasets confirm that the added modules successfully extend the capability of existing segmentators to yield highly competitive few-shot segmentation performances., Comment: 8 pages, 7 figures. arXiv admin note: text overlap with arXiv:2010.11437

Current-induced control of magnetization in ferromagnets using spin-orbit torque (SOT) has drawn attention as a new mechanism for fast and energy efficient magnetic memory devices. Energy-efficient spintronic devices require a spin-current source with a large SOT efficiency (${\xi}$) and electrical conductivity (${\sigma}$), and an efficient spin injection across a transparent interface. Herein, we use single crystals of the van der Waals (vdW) topological semimetal WTe$_2$ and vdW ferromagnet Fe$_3$GeTe$_2$ to satisfy the requirements in their all-vdW-heterostructure with an atomically sharp interface. The results exhibit values of ${\xi}{\approx}4.6$ and ${\sigma}{\approx}2.25{\times}10^5 {\Omega}^{-1} m^{-1}$ for WTe$_2$. Moreover, we obtain the significantly reduced switching current density of $3.90{\times}10^6 A/cm^2$ at 150 K, which is an order of magnitude smaller than those of conventional heavy-metal/ ferromagnet thin films. These findings highlight that engineering vdW-type topological materials and magnets offers a promising route to energy-efficient magnetization control in SOT-based spintronics., Comment: 19 pages, 4 figures

Few-shot learning allows machines to classify novel classes using only a few labeled samples. Recently, few-shot segmentation aiming at semantic segmentation on low sample data has also seen great interest. In this paper, we propose a learnable module for few-shot segmentation, the task-adaptive feature transformer (TAFT). TAFT linearly transforms task-specific high-level features to a set of task-agnostic features well-suited to the segmentation job. Using this task-conditioned feature transformation, the model is shown to effectively utilize the semantic information in novel classes to generate tight segmentation masks. The proposed TAFT module can be easily plugged into existing semantic segmentation algorithms to achieve few-shot segmentation capability with only a few added parameters. We combine TAFT with Deeplab V3+, a well-known segmentation architecture; experiments on the PASCAL-$5^i$ dataset confirm that this combination successfully adds few-shot learning capability to the segmentation algorithm, achieving the state-of-the-art few-shot segmentation performance in some key representative cases.

Learning novel concepts while preserving prior knowledge is a long-standing challenge in machine learning. The challenge gets greater when a novel task is given with only a few labeled examples, a problem known as incremental few-shot learning. We propose XtarNet, which learns to extract task-adaptive representation (TAR) for facilitating incremental few-shot learning. The method utilizes a backbone network pretrained on a set of base categories while also employing additional modules that are meta-trained across episodes. Given a new task, the novel feature extracted from the meta-trained modules is mixed with the base feature obtained from the pretrained model. The process of combining two different features provides TAR and is also controlled by meta-trained modules. The TAR contains effective information for classifying both novel and base categories. The base and novel classifiers quickly adapt to a given task by utilizing the TAR. Experiments on standard image datasets indicate that XtarNet achieves state-of-the-art incremental few-shot learning performance. The concept of TAR can also be used in conjunction with existing incremental few-shot learning methods; extensive simulation results in fact show that applying TAR enhances the known methods significantly., Comment: In Proceedings of the 37th International Conference on Machine Learning (ICML) 2020, Vienna, Austria, PMLR 119; *Equal contribution

We tackle a novel few-shot learning challenge, which we call few-shot semantic edge detection, aiming to localize crisp boundaries of novel categories using only a few labeled samples. We also present a Class-Agnostic Few-shot Edge detection Network (CAFENet) based on meta-learning strategy. CAFENet employs a semantic segmentation module in small-scale to compensate for lack of semantic information in edge labels. The predicted segmentation mask is used to generate an attention map to highlight the target object region, and make the decoder module concentrate on that region. We also propose a new regularization method based on multi-split matching. In meta-training, the metric-learning problem with high-dimensional vectors are divided into small subproblems with low-dimensional sub-vectors. Since there is no existing dataset for few-shot semantic edge detection, we construct two new datasets, FSE-1000 and SBD-$5^i$, and evaluate the performance of the proposed CAFENet on them. Extensive simulation results confirm the performance merits of the techniques adopted in CAFENet., Comment: 15 pages, 6 figures

Few-shot learning aims to handle previously unseen tasks using only a small amount of new training data. In preparing (or meta-training) a few-shot learner, however, massive labeled data are necessary. In the real world, unfortunately, labeled data are expensive and/or scarce. In this work, we propose a few-shot learner that can work well under the semi-supervised setting where a large portion of training data is unlabeled. Our method employs explicit task-conditioning in which unlabeled sample clustering for the current task takes place in a new projection space different from the embedding feature space. The conditioned clustering space is linearly constructed so as to quickly close the gap between the class centroids for the current task and the independent per-class reference vectors meta-trained across tasks. In a more general setting, our method introduces a concept of controlling the degree of task-conditioning for meta-learning: the amount of task-conditioning varies with the number of repetitive updates for the clustering space. Extensive simulation results based on the miniImageNet and tieredImageNet datasets show state-of-the-art semi-supervised few-shot classification performance of the proposed method. Simulation results also indicate that the proposed task-adaptive clustering shows graceful degradation with a growing number of distractor samples, i.e., unlabeled sample images coming from outside the candidate classes., Comment: 15 pages, 5 figures

In this paper, we consider hybrid uplink nonorthogonal multiple access (NOMA) that can support more users by exploiting the notion of power-domain NOMA. In hybrid uplink NOMA, we do not consider centralized power control as a base station (BS) needs instantaneous channel state information (CSI) of all users which leads to a high signaling overhead. Rather, each user is allowed to perform power control under fading in accordance with a truncated channel inversion power control policy. Due to the lack of coordination of centralized power control, users in the same resource block compete for access. To analyze users' behavior, evolutionary game can be considered so that each user can choose transmission strategies to maximize payoff in hybrid uplink NOMA with power control. Evolutionarily stable strategy (ESS) is characterized with fixed costs as well as costs that depend on channel realizations, and it is also shown that hybrid uplink NOMA can provide a higher throughput than orthogonal multiple access (OMA). To update the state in evolutionary game for hybrid uplink NOMA, the replicator dynamic equation is considered with two possible implementation methods., Comment: 11 pages; 8 figures; IEEE Trans. Comm

Handling previously unseen tasks after given only a few training examples continues to be a tough challenge in machine learning. We propose TapNets, neural networks augmented with task-adaptive projection for improved few-shot learning. Here, employing a meta-learning strategy with episode-based training, a network and a set of per-class reference vectors are learned across widely varying tasks. At the same time, for every episode, features in the embedding space are linearly projected into a new space as a form of quick task-specific conditioning. The training loss is obtained based on a distance metric between the query and the reference vectors in the projection space. Excellent generalization results in this way. When tested on the Omniglot, miniImageNet and tieredImageNet datasets, we obtain state of the art classification accuracies under various few-shot scenarios., Comment: in proceedings of the 36th International Conference on Machine Learning (ICML), Long Beach, PMLR 97:7115-7123, 2019

We propose a meta-learning algorithm utilizing a linear transformer that carries out null-space projection of neural network outputs. The main idea is to construct an alternative classification space such that the error signals during few-shot learning are quickly zero-forced on that space so that reliable classification on low data is possible. The final decision on a query is obtained utilizing a null-space-projected distance measure between the network output and reference vectors, both of which have been trained in the initial learning phase. Among the known methods with a given model size, our meta-learner achieves the best or near-best image classification accuracies with Omniglot and miniImageNet datasets., Comment: presented at 2018 NeurIPS (NIPS) Workshop on Meta-Learning (Montreal, Canada)