检索结果-内蒙古大学图书馆

Meta-ticket: finding optimal subnetworks for few-shot learning within randomly initialized neural networks 22

学校读者我要写书评

暂无评论

Meta-ticket: finding optimal subnetworks for few-shot learni...

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Daiki Chijiwa Shin'ya Yamaguchi Atsutoshi Kumagai Yasutoshi Ida NTT Computer and Data Science Laboratories NTT Corporation NTT Computer and Data Science Laboratories NTT Corporation and Kyoto University

ISBN: (纸本)9781713871088

Few-shot learning for neural networks (NNs) is an important problem that aims to train NNs with a few data. The main challenge is how to avoid overfitting since over-parameterized NNs can easily overfit to such small dataset. Previous work (e.g. MAML by Finn et al. 2017) tackles this challenge by meta-learning, which learns how to learn from a few data by using various tasks. On the other hand, one conventional approach to avoid overfitting is restricting hypothesis spaces by endowing sparse NN structures like convolution layers in computer vision. However, although such manually-designed sparse structures are sample-efficient for sufficiently large datasets, they are still insufficient for few-shot learning. Then the following questions naturally arise: (1) Can we find sparse structures effective for few-shot learning by meta-learning? (2) What benefits will it bring in terms of meta-generalization? In this work, we propose a novel meta-learning approach, called Meta-ticket, to find optimal sparse subnetworks for few-shot learning within randomly initialized NNs. We empirically validated that Meta-ticket successfully discover sparse subnetworks that can learn specialized features for each given task. Due to this task-wise adaptation ability, Meta-ticket achieves superior meta-generalization compared to MAML-based methods especially with large NNs.

关键词：

Few-shot learning for feature selection with hilbert-schmidt independence criterion 22

学校读者我要写书评

暂无评论

Few-shot learning for feature selection with hilbert-schmidt...

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Atsutoshi Kumagai Tomoharu Iwata Yasutoshi Ida Yasuhiro Fujiwara NTT Computer and Data Science Laboratories NTT Communication Science Laboratories

ISBN: (纸本)9781713871088

We propose a few-shot learning method for feature selection that can select relevant features given a small number of labeled instances. Existing methods require many labeled instances for accurate feature selection. However, sufficient instances are often unavailable. We use labeled instances in multiple related tasks to alleviate the lack of labeled instances in a target task. To measure the dependency between each feature and label, we use the Hilbert-Schmidt Independence Criterion, which is a kernel-based independence measure. By modeling the kernel functions with neural networks that take a few labeled instances in a task as input, we can encode the task- specific information to the kernels such that the kernels are appropriate for the task. Feature selection with such kernels is performed by using iterative optimization methods, in which each update step is obtained as a closed-form. This formulation enables us to directly and efficiently minimize the expected test error on features selected by a small number of labeled instances. We experimentally demonstrate that the proposed method outperforms existing feature selection methods.

关键词：

Selective Excitation of Superconducting Qubits with a Shared Control Line through Pulse Shaping

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Matsuda, R. Ohira, R. Sumida, T. Shiomi, H. Machino, A. Morisaka, S. Koike, K. Miyoshi, T. Kurimoto, Y. Sugita, Y. Ito, Y. Suzuki, Y. Spring, P.A. Wang, S. Tamate, S. Tabuchi, Y. Nakamura, Y. Ogawa, K. Negoro, M. Graduate School of Engineering Science Osaka University 1-3 Machikaneyama Toyonaka Osaka560-8531 Japan QuEL Inc. Daiwaunyu Building 3F 2-9-2 Owadamachi Hachioji Tokyo192-0045 Japan Center for Quantum Information and Quantum Biology Osaka University 1-2 Machikaneyama Toyonaka Osaka560-0043 Japan e-trees.Japan Inc. Daiwaunyu Building 2F 2-9-2 Owadamachi Hachioji Tokyo192-0045 Japan NTT Computer and Data Science Laboratories NTT Corporation Musashino180-8585 Japan RIKEN Center for Quantum Computing Saitama Wako351-0198 Japan Department of Applied Physics Graduate School of Engineering University of Tokyo 7-3-1 Hongo Bunkyo-ku Tokyo113-8656 Japan

In conventional architectures of superconducting quantum computers, each qubit is connected to its own control line, leading to a commensurate increase in the number of microwave lines as the system scales. Frequency-multiplexed qubit-control addresses this problem by enabling multiple qubits to share a single microwave line. However, it can cause unwanted excitation of non-target qubits, especially when the detuning between qubits is smaller than the pulse bandwidth. Here, we propose a selective-excitation-pulse (SEP) technique that suppresses unwanted excitations by shaping a drive pulse to create null points at non-target qubit frequencies. In a proof-of-concept experiment with three fixed-frequency transmon qubits, we demonstrate that the SEP technique achieves single-qubit gate fidelities comparable to those obtained with conventional Gaussian pulses while effectively suppressing unwanted excitations in non-target qubits. These results highlight the SEP technique as a promising tool for enhancing frequency-multiplexed qubit-control. © 2025, CC0.

关键词： Qubits

Few-shot learning for unsupervised feature selection

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Kumagai, Atsutoshi Iwata, Tomoharu Fujiwara, Yasuhiro NTT Computer and Data Science Laboratories NTT Communication Science Laboratories

We propose a few-shot learning method for unsupervised feature selection, which is a task to select a subset of relevant features in unlabeled data. Existing methods usually require many instances for feature selection. However, sufficient instances are often unavailable in practice. The proposed method can select a subset of relevant features in a target task given a few unlabeled target instances by training with unlabeled instances in multiple source tasks. Our model consists of a feature selector and decoder. The feature selector outputs a subset of relevant features taking a few unlabeled instances as input such that the decoder can reconstruct the original features of unseen instances from the selected ones. The feature selector uses the Concrete random variables to select features via gradient descent. To encode task-specific properties from a few unlabeled instances to the model, the Concrete random variables and decoder are modeled using permutation-invariant neural networks that take a few unlabeled instances as input. Our model is trained by minimizing the expected test reconstruction error given a few unlabeled instances that is calculated with datasets in source tasks. We experimentally demonstrate that the proposed method outperforms existing feature selection methods. Copyright © 2021, The Authors. All rights reserved.

关键词： Random variables

Fast Binary Network Hashing via Graph Clustering

学校读者我要写书评

暂无评论

Fast Binary Network Hashing via Graph Clustering

IEEE International Conference on Big data

作者： Yasuhiro Fujiwaray Masahiro Nakanoy Atsutoshi Kumagai Yasutoshi Idaz Akisato Kimura Naonori Ueda NTT Communication Science Laboratories NTT Computer and Data Science Laboratories

ISBN: (纸本)9781665480468

Network hashing converts each node of a graph into a compact binary code, and it is a useful graph analytics tool since it can reduce memory cost. INH-MF is a network hashing approach to factorize the high-order proximity matrix representing similarities between nodes. However, since it cuts small nonzero elements from the proximity matrix, it fails to effectively extract insights from the graph. Moreover, it incurs high memory and computational costs since the proximity matrix is large and dense. We propose Graph Clustering-based Network Hashing, a novel network hashing approach. To compute the proximities effectively, it uses the structural relationships between nodes and clusters obtained from a graph clustering approach. Moreover, it can efficiently compute hash codes from eigenvectors of the matrix corresponding to the graph Laplacian by using its low-rank property. Experiments show that it can more efficiently and effectively compute hash codes than previous approaches.

关键词： Laplace equations Costs Binary codes Big data Computational efficiency Matrix converters

Meta-learning for relative density-ratio estimation 21

学校读者我要写书评

暂无评论

Meta-learning for relative density-ratio estimation

Proceedings of the 35th International Conference on Neural Information Processing Systems

作者： Atsutoshi Kumagai Tomoharu Iwata Yasuhiro Fujiwara NTT Computer and Data Science Laboratories NTT Communication Science Laboratories

ISBN: (纸本)9781713845393

The ratio of two probability densities, called a density-ratio, is a vital quantity in machine learning. In particular, a relative density-ratio, which is a bounded extension of the density-ratio, has received much attention due to its stability and has been used in various applications such as outlier detection and dataset comparison. Existing methods for (relative) density-ratio estimation (DRE) require many instances from both densities. However, sufficient instances are often unavailable in practice. In this paper, we propose a meta-learning method for relative DRE, which estimates the relative density-ratio from a few instances by using knowledge in related datasets. Specifically, given two datasets that consist of a few instances, our model extracts the datasets' information by using neural networks and uses it to obtain instance embeddings appropriate for the relative DRE. We model the relative density-ratio by a linear model on the embedded space, whose global optimum solution can be obtained as a closed-form solution. The closed-form solution enables fast and effective adaptation to a few instances, and its differentiability enables us to train our model such that the expected test error for relative DRE can be explicitly minimized after adapting to a few instances. We empirically demonstrate the effectiveness of the proposed method by using three problems: relative DRE, dataset comparison, and outlier detection.

关键词：

META-LEARNING FOR OUT-OF-DISTRIBUTION DETECTION VIA DENSITY ESTIMATION IN LATENT SPACE

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Iwata, Tomoharu Kumagai, Atsutoshi NTT Communication Science Laboratories NTT Computer and Data Science Laboratories

Many neural network-based out-of-distribution (OoD) detection methods have been proposed. However, they require many training data for each target task. We propose a simple yet effective meta-learning method to detect OoD with small in-distribution data in a target task. With the proposed method, the OoD detection is performed by density estimation in a latent space. A neural network shared among all tasks is used to flexibly map instances in the original space to the latent space. The neural network is meta-learned such that the expected OoD detection performance is improved by using various tasks that are different from the target tasks. This meta-learning procedure enables us to obtain appropriate representations in the latent space for OoD detection. For density estimation, we use a Gaussian mixture model (GMM) with full covariance for each class. We can adapt the GMM parameters to in-distribution data in each task in a closed form by maximizing the likelihood. Since the closed form solution is differentiable, we can meta-learn the neural network efficiently with a stochastic gradient descent method by incorporating the solution into the meta-learning objective function. In experiments using six datasets, we demonstrate that the proposed method achieves better performance than existing meta-learning and OoD detection methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Stochastic systems

Meta-learning for relative density-ratio estimation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Kumagai, Atsutoshi Iwata, Tomoharu Fujiwara, Yasuhiro NTT Computer and Data Science Laboratories NTT Communication Science Laboratories

关键词： Anomaly detection

Resource-efficient generalized quantum subspace expansion

学校读者我要写书评

暂无评论

Physical Review Applied 2025年第5期23卷 054021-054021页

作者： Bo Yang Nobuyuki Yoshioka Hiroyuki Harada Shigeo Hakkaku Yuuki Tokunaga Hideaki Hakoshima Kaoru Yamamoto Suguru Endo LIP6 Sorbonne Université 4 Place Jussieu 75005 Paris France Graduate School of Information Science and Technology The University of Tokyo 7-3-1 Hongo Bunkyo-ku Tokyo 113-8656 Japan International Center for Elementary Particle Physics The University of Tokyo 7-3-1 Hongo Bunkyo-ku Tokyo 113-0033 Japan Theoretical Quantum Physics Laboratory RIKEN Cluster for Pioneering Research Wako-shi Saitama 351-0198 Japan JST PRESTO 4-1-8 Honcho Kawaguchi Saitama 332-0012 Japan Department of Applied Physics and Physico-Informatics Keio University Hiyoshi 3-14-1 Kohoku Yokohama 223-8522 Japan NTT Computer and Data Science Laboratories NTT Corporation 3-9-11 Midori-cho Musashino-shi Tokyo 180-8585 Japan Graduate School of Engineering Science Osaka University 1-3 Machikaneyama Toyonaka Osaka 560-8531 Japan Center for Quantum Information and Quantum Biology Osaka University 1-2 Machikaneyama Toyonaka Osaka 560-0043 Japan

Realizing practical quantum computing requires overcoming a number of computation errors and the limitation of device size, which have intensively been tackled by quantum error mitigation. As a unified approach of noise-agnostic quantum error mitigation, generalized quantum subspace expansion (GSE) has lately been proposed to be remarkably robust regarding stochastic and coherent errors, integrating quantum subspace expansion and virtual state purification. However, the requirement in GSE to perform entangled measurements between copies of the quantum states remains a significant drawback under the current situation of quantum devices with a restricted number of qubits and their connectivity. In this work, we propose “Dual-GSE,” a resource-efficient implementation of GSE to circumvent this overhead by constructing an ansatz of error-mitigated quantum states via dual-state purification without state copies. Remarkably, the proposed method can further simulate larger quantum systems beyond the size of available quantum hardware, achieved by a suitable ansatz construction inspired by the divide-and-conquer strategy that classically reintroduces the effect of entanglement. While classically forging the entanglement comes with additional cost, the total sampling overhead can be notably reduced by reuse of the same Pauli expectation values among divided-and-conquered subsystems. We comprehensively analyze the advantages and overhead of Dual-GSE and perform numerical simulations of the eight-qubit transverse-field Ising model under various setups. Our results demonstrate that Dual-GSE estimates the ground state energy with high accuracy under gate noise with low mitigation overhead and practical sampling cost.

关键词： Qubits