Search

The estimation and management of motor temperature are important for the continuous movements of robots. In this study, we propose an online learning method of thermal model parameters of motors for an accurate estimation of motor core temperature. Also, we propose a management method of motor core temperature using the updated model and anomaly detection method of motors. Finally, we apply this method to the muscles of the musculoskeletal humanoid and verify the ability of continuous movements., Comment: Accepted at IEEE Robotics and Automation Letters

This paper introduces LLM-jp, a cross-organizational project for the research and development of Japanese large language models (LLMs). LLM-jp aims to develop open-source and strong Japanese LLMs, and as of this writing, more than 1,500 participants from academia and industry are working together for this purpose. This paper presents the background of the establishment of LLM-jp, summaries of its activities, and technical reports on the LLMs developed by LLM-jp. For the latest activities, visit https://llm-jp.nii.ac.jp/en/.

In reinforcement learning (RL) with experience replay, experiences stored in a replay buffer influence the RL agent's performance. Information about the influence of these experiences is valuable for various purposes, such as identifying experiences that negatively influence poorly performing RL agents. One method for estimating the influence of experiences is the leave-one-out (LOO) method. However, this method is usually computationally prohibitive. In this paper, we present Policy Iteration with Turn-over Dropout (PIToD), which efficiently estimates the influence of experiences. We evaluate how accurately PIToD estimates the influence of experiences and its efficiency compared to LOO. We then apply PIToD to amend poorly performing RL agents, i.e., we use PIToD to estimate negatively influential experiences for the RL agents and to delete the influence of these experiences. We show that RL agents' performance is significantly improved via amendments with PIToD., Comment: Source code: https://github.com/TakuyaHiraoka/Which-Experiences-Are-Influential-for-RL-Agents

Library-based methods are known to be very effective for fast motion planning by adapting an experience retrieved from a precomputed library. This article presents CoverLib, a principled approach for constructing and utilizing such a library. CoverLib iteratively adds an experience-classifier-pair to the library, where each classifier corresponds to an adaptable region of the experience within the problem space. This iterative process is an active procedure, as it selects the next experience based on its ability to effectively cover the uncovered region. During the query phase, these classifiers are utilized to select an experience that is expected to be adaptable for a given problem. Experimental results demonstrate that CoverLib effectively mitigates the trade-off between plannability and speed observed in global (e.g. sampling-based) and local (e.g. optimization-based) methods. As a result, it achieves both fast planning and high success rates over the problem domain. Moreover, due to its adaptation-algorithm-agnostic nature, CoverLib seamlessly integrates with various adaptation methods, including nonlinear programming-based and sampling-based algorithms.

The stability of persistent homology has led to wide applications of the persistence diagram as a trusted topological descriptor in the presence of noise. However, with the increasing demand for high-dimension and low-sample-size data processing in modern science, it is questionable whether persistence diagrams retain their reliability in the presence of high-dimensional noise. This work aims to study the reliability of persistence diagrams in the high-dimension low-sample-size data setting. By analyzing the asymptotic behavior of persistence diagrams for high-dimensional random data, we show that persistence diagrams are no longer reliable descriptors of low-sample-size data under high-dimensional noise perturbations. We refer to this loss of reliability of persistence diagrams in such data settings as the curse of dimensionality on persistence diagrams. Next, we investigate the possibility of using normalized principal component analysis as a method for reducing the dimensionality of the high-dimensional observed data to resolve the curse of dimensionality. We show that this method can mitigate the curse of dimensionality on persistence diagrams. Our results shed some new light on the challenges of processing high-dimension low-sample-size data by persistence diagrams and provide a starting point for future research in this area.

We explore threshold vocabulary trimming in Byte-Pair Encoding subword tokenization, a postprocessing step that replaces rare subwords with their component subwords. The technique is available in popular tokenization libraries but has not been subjected to rigorous scientific scrutiny. While the removal of rare subwords is suggested as best practice in machine translation implementations, both as a means to reduce model size and for improving model performance through robustness, our experiments indicate that, across a large space of hyperparameter settings, vocabulary trimming fails to improve performance, and is even prone to incurring heavy degradation., Comment: 15 pages

Large language models (LLMs) have increased interest in vision language models (VLMs), which process image-text pairs as input. Studies investigating the visual understanding ability of VLMs have been proposed, but such studies are still preliminary because existing datasets do not permit a comprehensive evaluation of the fine-grained visual linguistic abilities of VLMs across multiple languages. To further explore the strengths of VLMs, such as GPT-4V \cite{openai2023GPT4}, we developed new datasets for the systematic and qualitative analysis of VLMs. Our contribution is four-fold: 1) we introduced nine vision-and-language (VL) tasks (including object recognition, image-text matching, and more) and constructed multilingual visual-text datasets in four languages: English, Japanese, Swahili, and Urdu through utilizing templates containing \textit{questions} and prompting GPT4-V to generate the \textit{answers} and the \textit{rationales}, 2) introduced a new VL task named \textit{unrelatedness}, 3) introduced rationales to enable human understanding of the VLM reasoning process, and 4) employed human evaluation to measure the suitability of proposed datasets for VL tasks. We show that VLMs can be fine-tuned on our datasets. Our work is the first to conduct such analyses in Swahili and Urdu. Also, it introduces \textit{rationales} in VL analysis, which played a vital role in the evaluation.

Do pretrained language models have knowledge regarding the surface information of tokens? We examined the surface information stored in word or subword embeddings acquired by pretrained language models from the perspectives of token length, substrings, and token constitution. Additionally, we evaluated the ability of models to generate knowledge regarding token surfaces. We focused on 12 pretrained language models that were mainly trained on English and Japanese corpora. Experimental results demonstrate that pretrained language models have knowledge regarding token length and substrings but not token constitution. Additionally, the results imply that there is a bottleneck on the decoder side in terms of effectively utilizing acquired knowledge.

Transferring human motion skills to humanoid robots remains a significant challenge. In this study, we introduce a Wasserstein adversarial imitation learning system, allowing humanoid robots to replicate natural whole-body locomotion patterns and execute seamless transitions by mimicking human motions. First, we present a unified primitive-skeleton motion retargeting to mitigate morphological differences between arbitrary human demonstrators and humanoid robots. An adversarial critic component is integrated with Reinforcement Learning (RL) to guide the control policy to produce behaviors aligned with the data distribution of mixed reference motions. Additionally, we employ a specific Integral Probabilistic Metric (IPM), namely the Wasserstein-1 distance with a novel soft boundary constraint to stabilize the training process and prevent mode collapse. Our system is evaluated on a full-sized humanoid JAXON in the simulator. The resulting control policy demonstrates a wide range of locomotion patterns, including standing, push-recovery, squat walking, human-like straight-leg walking, and dynamic running. Notably, even in the absence of transition motions in the demonstration dataset, robots showcase an emerging ability to transit naturally between distinct locomotion patterns as desired speed changes.

We performed time-resolved pump--probe measurements using rare-earth iron garnet \ce{Gd3/2Yb1/2BiFe5O12} as a two-sublattice ferrimagnet. We measured the initial phases of the magnetic resonance modes below and above the magnetization compensation temperature to clarify the sublattice selectivity of the inverse Faraday effect in ferrimagnets. A comparison of the time evolution of magnetization estimated using the equations of motion revealed that the inverse Faraday effect occurring in ferrimagnetic materials has sublattice selectivity. This is in striking contrast to antiferromagnets, in which the inverse Faraday effect acts on each sublattice identically. The initial phase analysis can be applied to other ferrimagnets with compensation temperatures., Comment: 4 pages, 5 figures

Reinforcement learning (RL) methods with a high replay ratio (RR) and regularization have gained interest due to their superior sample efficiency. However, these methods have mainly been developed for dense-reward tasks. In this paper, we aim to extend these RL methods to sparse-reward goal-conditioned tasks. We use Randomized Ensemble Double Q-learning (REDQ) (Chen et al., 2021), an RL method with a high RR and regularization. To apply REDQ to sparse-reward goal-conditioned tasks, we make the following modifications to it: (i) using hindsight experience replay and (ii) bounding target Q-values. We evaluate REDQ with these modifications on 12 sparse-reward goal-conditioned tasks of Robotics (Plappert et al., 2018), and show that it achieves about $2 \times$ better sample efficiency than previous state-of-the-art (SoTA) RL methods. Furthermore, we reconsider the necessity of specific components of REDQ and simplify it by removing unnecessary ones. The simplified REDQ with our modifications achieves $\sim 8 \times$ better sample efficiency than the SoTA methods in 4 Fetch tasks of Robotics., Comment: Source code: https://github.com/TakuyaHiraoka/Efficient-SRGC-RL-with-a-High-RR-and-Regularization Demo video: https://drive.google.com/file/d/1UHd7JVPCwFLNFhy1QcycQfwU_nll_yII/view?usp=drive_link

A fundamental challenge in multiparameter persistent homology is the absence of a complete and discrete invariant. To address this issue, we propose an enhanced framework that realizes a holistic understanding of a fully commutative quiver's representation via synthesizing interpretations obtained from intervals. Additionally, it provides a mechanism to tune the balance between approximation resolution and computational complexity. This framework is evaluated on commutative ladders of both finite-type and infinite-type. For the former, we discover an efficient method for the indecomposable decomposition leveraging solely one-parameter persistent homology. For the latter, we introduce a new invariant that reveals persistence in the second parameter by connecting two standard persistence diagrams using interval approximations. We subsequently present several models for constructing commutative ladder filtrations, offering fresh insights into random filtrations and demonstrating our toolkit's effectiveness in analyzing the topology of materials.

Imitation learning has been actively studied in recent years. In particular, skill acquisition by a robot with a fixed body, whose root link position and posture and camera angle of view do not change, has been realized in many cases. On the other hand, imitation of the behavior of robots with floating links, such as humanoid robots, is still a difficult task. In this study, we develop an imitation learning system using a biped robot with a floating link. There are two main problems in developing such a system. The first is a teleoperation device for humanoids, and the second is a control system that can withstand heavy workloads and long-term data collection. For the first point, we use the whole body control device TABLIS. It can control not only the arms but also the legs and can perform bilateral control with the robot. By connecting this TABLIS with the high-power humanoid robot JAXON, we construct a control system for imitation learning. For the second point, we will build a system that can collect long-term data based on posture optimization, and can simultaneously move the robot's limbs. We combine high-cycle posture generation with posture optimization methods, including whole-body joint torque minimization and contact force optimization. We designed an integrated system with the above two features to achieve various tasks through imitation learning. Finally, we demonstrate the effectiveness of this system by experiments of manipulating flexible fabrics such that not only the hands but also the head and waist move simultaneously, manipulating objects using legs characteristic of humanoids, and lifting heavy objects that require large forces., Comment: accepted at IROS2023

Node2vec is a graph embedding method that learns a vector representation for each node of a weighted graph while seeking to preserve relative proximity and global structure. Numerical experiments suggest Node2vec struggles to recreate the topology of the input graph. To resolve this we introduce a topological loss term to be added to the training loss of Node2vec which tries to align the persistence diagram (PD) of the resulting embedding as closely as possible to that of the input graph. Following results in computational optimal transport, we carefully adapt entropic regularization to PD metrics, allowing us to measure the discrepancy between PDs in a differentiable way. Our modified loss function can then be minimized through gradient descent to reconstruct both the geometry and the topology of the input graph. We showcase the benefits of this approach using demonstrative synthetic examples., Comment: For associated repository, see https://github.com/killianfmeehan/topological_node2vec

When a fraction of a population becomes immune to an infectious disease, the population-wide infection risk decreases nonlinearly due to collective protection known as herd immunity. Studies based on mean-field models suggest that natural infection in a heterogeneous population may induce herd immunity more efficiently than homogeneous immunization. Here, we use network epidemic models to show that the opposite can also be the case. We identify two competing mechanisms driving disease-induced herd immunity in networks: the high density of immunity among socially active individuals enhances the herd immunity effect, while the topological localization of immune individuals weakens it. The effect of localization is stronger in networks embedded in low-dimensional space, which can make disease-induced immunity less effective than random immunization. Our results highlight the role of networks in shaping herd immunity and call for careful examination of model predictions that inform public health policies., Comment: Main text: 11 pages, 4 figures. Supplementary Materials: 8 pages, 2 figures

We explored quinary body-centered cubic (bcc) high-entropy alloy (HEA) superconductors with valence electron concentrations (VECs) ranging from 4.6 to 5.0, a domain that has received limited attention in prior research. Our search has led to the discovery of new bcc Ti-Hf-Nb-Ta-Re superconducting alloys, which exhibit an interesting phenomenon of phase segregation into two bcc phases with slightly different chemical compositions, as the VEC increases. The enthalpy of the formation of each binary compound explains the phase segregation. All the alloys investigated were categorized as type-II superconductors, with superconducting critical temperatures ($T_\mathrm{c}$) ranging from 3.25 K to 4.38 K. We measured the Vickers microhardness, which positively correlated with the Debye temperature, and compared it with the hardness values of other bcc HEA superconductors. Our results indicate that $T_\mathrm{c}$ systematically decreases with an increase in hardness beyond a threshold of approximately 350 HV. Additionally, we plotted $T_\mathrm{c}$ vs. VEC for representative quinary bcc HEAs. The plot revealed the asymmetric VEC dependence. The correlation between the hardness and $T_\mathrm{c}$, as well as the asymmetric dependence of $T_\mathrm{c}$ on VEC can be attributed to the simultaneous effects of the electronic density of states at the Fermi level and electron-phonon coupling under the uncertainty principle, especially in the higher VEC region., Comment: to appear in Journal of Alloys and Metallurgical Systems

Xenon dual-phase time projections chambers (TPCs) have proven to be a successful technology in studying physical phenomena that require low-background conditions. With 40t of liquid xenon (LXe) in the TPC baseline design, DARWIN will have a high sensitivity for the detection of particle dark matter, neutrinoless double beta decay ($0\nu\beta\beta$), and axion-like particles (ALPs). Although cosmic muons are a source of background that cannot be entirely eliminated, they may be greatly diminished by placing the detector deep underground. In this study, we used Monte Carlo simulations to model the cosmogenic background expected for the DARWIN observatory at four underground laboratories: Laboratori Nazionali del Gran Sasso (LNGS), Sanford Underground Research Facility (SURF), Laboratoire Souterrain de Modane (LSM) and SNOLAB. We determine the production rates of unstable xenon isotopes and tritium due to muon-included neutron fluxes and muon-induced spallation. These are expected to represent the dominant contributions to cosmogenic backgrounds and thus the most relevant for site selection.

Level 3 automated driving systems (ADS) have attracted significant attention and are being commercialized. A Level 3 ADS prompts the driver to take control by requesting to intervene (RtI) when its operational design domain (ODD) or system limitations are exceeded. However, complex traffic situations may lead drivers to perceive multiple potential triggers of RtI simultaneously, causing hesitation or confusion during take-over. Therefore, drivers must clearly understand the ADS's system limitations to understand the triggers of RtI and ensure safe take-over. In this study, we propose a voice-based HMI for providing RtI trigger cues to help drivers understand ADS's system limitations. The results of a between-group experiment using a driving simulator showed that incorporating effective trigger cues into the RtI enabled drivers to comprehend the ADS's system limitations better and reduce collisions. It also improved the subjective evaluations of drivers, such as the comprehensibility of system limitations, hesitation in response to RtI, and acceptance of ADS behaviors when encountering RtI while using the ADS. Therefore, enhanced comprehension resulting from trigger cues is essential for promoting a safer and better user experience using ADS during RtI.

We have synthesized new Tsai-type Yb-based intermediate-valence approximant crystals (ACs) with different degree of approximation to quasicrystal, Zn--Au--Yb 1/1 and 2/1 AC, and studied the external pressure effect on their Yb mean-valence $\nu$. Whereas 1/1 AC distinctly exhibits a first-order-like jump in $\nu$ at a transition pressure $P_{\rm v}$, 2/1 AC only shows an indistinct anomaly at $P_{\rm v}$. We have also studied the pressure dependence of the $\nu$ of Au--Al--Yb 1/1 AC, which is a prototypal AC exhibiting pressure-induced quantum criticality. It shows a continuous valence anomaly at a critical pressure $P_{\rm c}$ where the magnetic susceptibility diverges toward zero temperature, in contrast to the valence jump in the Zn--Au--Yb 1/1 AC. These results are discussed based on a theoretical model of quantum critical valence fluctuation.

Recently, methods for learning diverse skills to generate various behaviors without external rewards have been actively studied as a form of unsupervised reinforcement learning. However, most of the existing methods learn a finite number of discrete skills, and thus the variety of behaviors that can be exhibited with the learned skills is limited. In this paper, we propose a novel method for learning potentially an infinite number of different skills, which is named discovery of continuous skills on a sphere (DISCS). In DISCS, skills are learned by maximizing mutual information between skills and states, and each skill corresponds to a continuous value on a sphere. Because the representations of skills in DISCS are continuous, infinitely diverse skills could be learned. We examine existing methods and DISCS in the MuJoCo Ant robot control environments and show that DISCS can learn much more diverse skills than the other methods., Comment: 14 pages, 12 figures

Is preferred tokenization for humans also preferred for machine-learning (ML) models? This study examines the relations between preferred tokenization for humans (appropriateness and readability) and one for ML models (performance on an NLP task). The question texts of the Japanese commonsense question-answering dataset are tokenized with six different tokenizers, and the performances of human annotators and ML models were compared. Furthermore, we analyze relations among performance of answers by human and ML model, the appropriateness of tokenization for human, and response time to questions by human. This study provides a quantitative investigation result that shows that preferred tokenizations for humans and ML models are not necessarily always the same. The result also implies that existing methods using language models for tokenization could be a good compromise both for human and ML models.

This paper proposes a method to optimize tokenization for the performance improvement of already trained downstream models. Our method generates tokenization results attaining lower loss values of a given downstream model on the training data for restricting vocabularies and trains a tokenizer reproducing the tokenization results. Therefore, our method can be applied to variety of tokenization methods, while existing work cannot due to the simultaneous learning of the tokenizer and the downstream model. This paper proposes an example of the BiLSTM-based tokenizer with vocabulary restriction, which can capture wider contextual information for the tokenization process than non-neural-based tokenization methods used in existing work. Experimental results on text classification in Japanese, Chinese, and English text classification tasks show that the proposed method improves performance compared to the existing methods for tokenization optimization.

In reinforcement learning (RL) with experience replay, experiences stored in a replay buffer influence the RL agent's performance. Information about the influence is valuable for various purposes, including experience cleansing and analysis. One method for estimating the influence of individual experiences is agent comparison, but it is prohibitively expensive when there is a large number of experiences. In this paper, we present PI+ToD as a method for efficiently estimating the influence of experiences. PI+ToD is a policy iteration that efficiently estimates the influence of experiences by utilizing turn-over dropout. We demonstrate the efficiency of PI+ToD with experiments in MuJoCo environments., Comment: The paper is withdrawn because an error that affects the main results of the experiments has been found

The objective of this article is to investigate the asymptotic behavior of the persistence diagrams of a random cubical filtration as the window size tends to infinity. Here, a random cubical filtration is an increasing family of random cubical sets, which are the union of randomly generated higher-dimensional unit cubes with integer coordinates in a Euclidean space. We first prove the strong law of large numbers for the persistence diagrams, inspired by the work of Hiraoka, Shirai, and Trinh, where the persistence diagram of a filtration of random geometric complexes is considered. As opposed to prior papers treating limit theorems for persistence diagrams, the present article aims to further study the large deviation behavior of persistence diagrams. We prove a large deviation principle for the persistence diagrams of a class of random cubical filtrations, and show that the rate function is given as the Fenchel--Legendre transform of the limiting logarithmic moment generating function. In the proof, we also establish a general method of lifting a large deviation principle for the tuples of persistent Betti numbers to persistence diagrams for broad application., Comment: 31 pages

We present a subword regularization method for WordPiece, which uses a maximum matching algorithm for tokenization. The proposed method, MaxMatch-Dropout, randomly drops words in a search using the maximum matching algorithm. It realizes finetuning with subword regularization for popular pretrained language models such as BERT-base. The experimental results demonstrate that MaxMatch-Dropout improves the performance of text classification and machine translation tasks as well as other subword regularization methods. Moreover, we provide a comparative analysis of subword regularization methods: subword regularization with SentencePiece (Unigram), BPE-Dropout, and MaxMatch-Dropout., Comment: Accepted to appear at COLING2022

We report the first study of electrical resistivity, magnetization, and specific heat on YbCo$_2$. The measurements on a single-phased sample of YbCo$_2$ bring no evidence of magnetic ordering down to 0.3 K in a zero magnetic field. The manifestations of low Kondo temperature are observed. The specific heat value divided by temperature, C/T, keeps increasing logarithmically beyond 7 J/mol.K2 with decreasing temperature down to 0.3 K without no sign of magnetic ordering, suggesting a very large electronic specific heat. Analysis of the magnetic specific heat indicates that the large portion of the low-temperature specific heat is not explained simply by the low Kondo temperature but is due to the strong intersite magnetic correlation in both the 3d and 4f electrons. Temperature-dependent measurements under static magnetic fields up to 7 T are carried out, which show the evolution of field-induced transition above 2 T. The transition temperature increases with increasing field, pointing to a ferromagnetic character. The extrapolation of the transition temperature to zero field suggests that YbCo$_2$ is in the very proximity of the quantum critical point. These results indicate that in the unique case of YbCo$_2$, the itinerant electron magnetism of Co 3d-electrons and the Kondo effect within the vicinity of quantum criticality of Yb 4f-local moments can both play a role.

Subword regularizations use multiple subword segmentations during training to improve the robustness of neural machine translation models. In previous subword regularizations, we use multiple segmentations in the training process but use only one segmentation in the inference. In this study, we propose an inference strategy to address this discrepancy. The proposed strategy approximates the marginalized likelihood by using multiple segmentations including the most plausible segmentation and several sampled segmentations. Because the proposed strategy aggregates predictions from several segmentations, we can regard it as a single model ensemble that does not require any additional cost for training. Experimental results show that the proposed strategy improves the performance of models trained with subword regularization in low-resource machine translation tasks., Comment: Findings of ACL 2022

We study how the herd immunity threshold and the expected epidemic size depend on homophily with respect to vaccine adoption. We find that the presence of homophily considerably increases the critical vaccine coverage needed for herd immunity and that strong homophily can push the threshold entirely out of reach. The epidemic size monotonically increases as a function of homophily strength for a perfect vaccine, while it is maximized at a nontrivial level of homophily when the vaccine efficacy is limited. Our results highlight the importance of vaccination homophily in epidemic modeling., Comment: 12 pages, 9 figures

High-entropy-alloy (HEA) superconductors are a new class of disordered superconductors. In this study, we investigate the robustness of superconducting states in HEA-type metal telluride (MTe; M = Ag, In, Sn, Pb, Bi) under high pressure. PbTe exhibits a structural transition from a NaCl-type to an orthorhombic Pnma structure at low pressures, and further transitions to a CsCl-type structure at high pressures. When the superconductivity of the CsCl-type PbTe is observed, it is found that its superconducting transition temperature (Tc) decreases with pressure. However, in the HEA-type AgInSnPbBiTe5, Tc is almost independent of pressure, for pressures ranging from 13.0 to 35.1 GPa. This trend is quite similar to that observed in an HEA superconductor (TaNb)0.67(HfZrTi)0.33, which shows that the robustness of superconductivity to external pressure is a universal feature in HEA-type superconductors. To clarify the effects of the modification of the configurational entropy of mixing on the crystal structure, superconducting states, and electronic structure of MTe, electrical resistance measurements, synchrotron X-ray diffraction, and synchrotron X-ray absorption spectroscopy with partial fluorescence mode (PFY-XAS) for three MTe polycrystalline samples of PbTe, AgPbBiTe3, and AgInSnPbBiTe5 with different configurational entropies of mixing at the M site were performed., Comment: 22 pages, 4 figures, supporting information

This paper introduces persistent homology, which is a powerful tool to characterize the shape of data using the mathematical concept of topology. We explain the fundamental idea of persistent homology from scratch using some examples. We also review some applications of persistent homology to materials researches and software for persistent homology data analysis. HomCloud, one of persistent homology software, is especially featured in this paper.

Randomized ensembled double Q-learning (REDQ) (Chen et al., 2021b) has recently achieved state-of-the-art sample efficiency on continuous-action reinforcement learning benchmarks. This superior sample efficiency is made possible by using a large Q-function ensemble. However, REDQ is much less computationally efficient than non-ensemble counterparts such as Soft Actor-Critic (SAC) (Haarnoja et al., 2018a). To make REDQ more computationally efficient, we propose a method of improving computational efficiency called DroQ, which is a variant of REDQ that uses a small ensemble of dropout Q-functions. Our dropout Q-functions are simple Q-functions equipped with dropout connection and layer normalization. Despite its simplicity of implementation, our experimental results indicate that DroQ is doubly (sample and computationally) efficient. It achieved comparable sample efficiency with REDQ, much better computational efficiency than REDQ, and comparable computational efficiency with that of SAC., Comment: ICLR 2022. Source code: https://github.com/TakuyaHiraoka/Dropout-Q-Functions-for-Doubly-Efficient-Reinforcement-Learning Poster: https://drive.google.com/file/d/1_JSuwlUsMjzo6zRaAIcXXj3__AmOvu2t/view?usp=sharing Slides: https://drive.google.com/file/d/1ecq9SQ2KSNpfeblCkr6TYPz5gRk_Y4S8/view?usp=sharing

In current times, the importance of online hotel review sites has become more and more apparent. Users of these sites reference of reviews strongly influences their purchase behavior and as such, reviews are important to companies and researchers alike. The majority of review sites offer both text reviews and numerical hotel ratings, and both information sources are widely used by researchers as a representation of a customer's sentiment and opinion. However, an opinion is a difficult concept to measure, and as such, depending on the relation these two sources have, it would be apparent whether or not it is safe to consider them equally in research. In this study we utilize an entropy-based Support Vector Machine to classify positive and negative sentiments in hotel reviews from the site Ctrip, then calculating the ratio of positive and negative sentiment in each review and examine their correlation with said review's rating score using Spearman and Kendall Correlation coefficients and Maximal Information Coefficient (MIC)., Comment: Translation of the original in Japanese

A polymorph of RuI$_3$ synthesized under high pressure was found to have a two-layered honeycomb structure. The resistivity of RuI$_3$ exhibits a semimetallic behavior, in contrast to insulating properties in $\alpha$-RuCl$_3$. In addition, Pauli paramagnetic behavior was observed in the temperature dependence of a magnetic susceptibility and a nuclear spin-lattice relaxation rate 1/$T_1$. The band structure calculations indicate that contribution of the I 5$p$ components to the low-energy $t_\mathrm{2g}$ bands effectively decreases Coulomb repulsion, leading to semimetallic properties. The physical properties also suggest strong electron correlations in RuI$_3$., Comment: 4 Figures

Since culture influences expectations, perceptions, and satisfaction, a cross-culture study is necessary to understand the differences between Japan's biggest tourist populations, Chinese and Western tourists. However, with ever-increasing customer populations, this is hard to accomplish without extensive customer base studies. There is a need for an automated method for identifying these expectations at a large scale. For this, we used a data-driven approach to our analysis. Our study analyzed their satisfaction factors comparing soft attributes, such as service, with hard attributes, such as location and facilities, and studied different price ranges. We collected hotel reviews and extracted keywords to classify the sentiment of sentences with an SVC. We then used dependency parsing and part-of-speech tagging to extract nouns tied to positive adjectives. We found that Chinese tourists consider room quality more than hospitality, whereas Westerners are delighted more by staff behavior. Furthermore, the lack of a Chinese-friendly environment for Chinese customers and cigarette smell for Western ones can be disappointing factors of their stay. As one of the first studies in the tourism field to use the high-standard Japanese hospitality environment for this analysis, our cross-cultural study contributes to both the theoretical understanding of satisfaction and suggests practical applications and strategies for hotel managers., Comment: Published Online at: https://link.springer.com/article/10.1007%2Fs40558-021-00203-8

A complex interplay of different energy scales involving Coulomb repulsion, spin-orbit coupling and Hund's coupling energy in two-dimensional (2D) van der Waals (vdW) material produces novel emerging physical state. For instance, ferromagnetism in vdW charge transfer insulator CrGeTe$_3$, that provides a promising platform to simultaneously manipulate the magnetic and electrical properties for potential device implementation using few layers thick materials. Here, we show a continuous tuning of magnetic and electrical properties of CrGeTe$_3$ single crystal using pressure. With application of pressure, CrGeTe$_3$ transforms from a FM insulator with Curie temperature, $T_{\rm{C}} \sim $ 66 K at ambient condition to a correlated 2D Fermi metal with $T_{\rm{C}}$ exceeding $\sim$ 250 K. Notably, absence of an accompanying structural distortion across the insulator-metal transition (IMT) suggests that the pressure induced modification of electronic ground states are driven by electronic correlation furnishing a rare example of bandwidth-controlled IMT in a vdW material., Comment: 6 pages, 4 figures

The Thinking Wave is an ongoing development of visualization concepts showing the real-time effort and confidence of semi-autonomous vehicle (AV) systems. Offering drivers access to this information can inform their decision making, and enable them to handle the situation accordingly and takeover when necessary. Two different visualizations have been designed, Concept one, Tidal, demonstrates the AV systems effort through intensified activity of a simple graphic which fluctuates in speed and frequency. Concept two, Tandem, displays the effort of the AV system as well as the handling dynamic and shared responsibility between the driver and the vehicle system. Working collaboratively with mobility research teams at the University of Tokyo, we are prototyping and refining the Thinking Wave and its embodiments as we work towards building a testable version integrated into a driving simulator. The development of the thinking wave aims to calibrate trust by increasing the drivers knowledge and understanding of vehicle handling capacity. By enabling transparent communication of the AV systems capacity, we hope to empower AV-skeptic drivers and keep over-trusting drivers on alert in the case of an emergency takeover situation, in order to create a safer autonomous driving experience., Comment: 8 pages, 11 figures, 2021 IEEE Intelligent Vehicles Symposium (IV21)

Trust is a multilayered concept with critical relevance when it comes to introducing new technologies. Understanding how humans will interact with complex vehicle systems and preparing for the functional, societal and psychological aspects of autonomous vehicles' entry into our cities is a pressing concern. Design tools can help calibrate the adequate and affordable level of trust needed for a safe and positive experience. This study focuses on passenger interactions capable of enhancing the system trustworthiness and data accuracy in future shared public transportation., Comment: 4 pages, 5 figures, IV 2021 Nagoya, Trust Calibration Workshop

Since traditional tokenizers are isolated from a downstream task and model, they cannot output an appropriate tokenization depending on the task and model, although recent studies imply that the appropriate tokenization improves the performance. In this paper, we propose a novel method to find an appropriate tokenization to a given downstream model by jointly optimizing a tokenizer and the model. The proposed method has no restriction except for using loss values computed by the downstream model to train the tokenizer, and thus, we can apply the proposed method to any NLP task. Moreover, the proposed method can be used to explore the appropriate tokenization for an already trained model as post-processing. Therefore, the proposed method is applicable to various situations. We evaluated whether our method contributes to improving performance on text classification in three languages and machine translation in eight language pairs. Experimental results show that our proposed method improves the performance by determining appropriate tokenizations., Comment: Accepted at ACL-IJCNLP 2021 Findings

We evaluate the efficiency of various heuristic strategies for allocating vaccines against COVID-19 and compare them to strategies found using optimal control theory. Our approach is based on a mathematical model which tracks the spread of disease among different age groups and across different geographical regions, and we introduce a method to combine age-specific contact data to geographical movement data. As a case study, we model the epidemic in the population of mainland Finland utilizing mobility data from a major telecom operator. Our approach allows to determine which geographical regions and age groups should be targeted first in order to minimize the number of deaths. In the scenarios that we test, we find that distributing vaccines demographically and in an age-descending order is not optimal for minimizing deaths and the burden of disease. Instead, more lives could potentially be saved by using strategies which emphasize high-incidence regions and distribute vaccines in parallel to multiple age groups. The level of emphasis that high-incidence regions should be given depends on the overall transmission rate in the population. This observation highlights the importance of updating the vaccination strategy when the effective reproduction number changes due to the general contact patterns changing and new virus variants entering., Comment: Revision

We propose an efficient framework with compatibility between normal printing and printing with special color inks in this paper. Special color inks can be used for printing to represent some particular colors and specific optical properties, which are difficult to express using only CMYK inks. Special color layers are required in addition to the general color layer for printing with special color inks. We introduce a reversible data hiding (RDH) method to embed the special color layers into the general color layer without visible artifacts. The proposed method can realize both normal printing and printing with special color inks by using a single layer. Our experimental results show that the quality of the marked image is virtually identical to that of the original image, i.e., the general color layer., Comment: 8 pages

This paper studies the magnitude homology of graphs focusing mainly on the relationship between its diagonality and the girth. Magnitude and magnitude homology are formulations of the Euler characteristic and the corresponding homology, respectively, for finite metric spaces, first introduced by Leinster and Hepworth-Willerton. Several authors study them restricting to graphs with path metric, and some properties which are similar to the ordinary homology theory have come to light. However, the whole picture of their behavior is still unrevealed, and it is expected that they catch some geometric properties of graphs. In this article, we show that the girth of graphs partially determines magnitude homology, that is, the larger girth a graph has, the more homologies near the diagonal part vanish. Furthermore, applying this result to a typical random graph, we investigate how the diagonality of graphs varies statistically as the edge density increases. In particular, we show that there exists a phase transition phenomenon for the diagonality., Comment: 21 pages, 5 figures, a reference added

We studied the injection-locking properties of a resonant-tunneling-diode terahertz oscillator in the small-signal injection regime with a frequency-stabilized continuous THz wave. The linewidth of the emission spectrum dramatically decreased to less than 120 mHz (HWHM) from 4.4 MHz in the free running state as a result of the injection locking. We experimentally determined the amplitude of injection voltage at the antenna caused by the injected THz wave. The locking range was proportional to the injection amplitude and consistent with Adler's model. As increasing the injection amplitude, we observed decrease of the noise component in the power spectrum, which manifests the free-running state, and alternative increase of the injection-locked component. The noise component and the injection-locked component had the same power at the threshold injection amplitude as small as $5\times10^{-4}$ of the oscillation amplitude. This threshold behavior can be qualitatively explained by Maffezzoni's model of noise reduction in general limit-cycle oscillators., Comment: The following article has been submitted to APL Photonics

Meta-reinforcement learning (RL) addresses the problem of sample inefficiency in deep RL by using experience obtained in past tasks for a new task to be solved. However, most meta-RL methods require partially or fully on-policy data, i.e., they cannot reuse the data collected by past policies, which hinders the improvement of sample efficiency. To alleviate this problem, we propose a novel off-policy meta-RL method, embedding learning and evaluation of uncertainty (ELUE). An ELUE agent is characterized by the learning of a feature embedding space shared among tasks. It learns a belief model over the embedding space and a belief-conditional policy and Q-function. Then, for a new task, it collects data by the pretrained policy, and updates its belief based on the belief model. Thanks to the belief update, the performance can be improved with a small amount of data. In addition, it updates the parameters of the neural networks to adjust the pretrained relationships when there are enough data. We demonstrate that ELUE outperforms state-of-the-art meta RL methods through experiments on meta-RL benchmarks., Comment: 14pages

The origin of non-Poissonian or bursty temporal patterns observed in various datasets for human social dynamics has been extensively studied, yet its understanding still remains incomplete. Considering the fact that humans are social beings, a fundamental question arises: Is the bursty human dynamics dominated by individual characteristics or by interaction between individuals? In this paper we address this question by analyzing the Wikipedia edit history to see how spontaneous individual editors are in initiating bursty periods of editing, i.e., individual-driven burstiness, and to what extent such editors' behaviors are driven by interaction with other editors in those periods, i.e., interaction-driven burstiness. We quantify the degree of initiative (DoI) of an editor of interest in each Wikipedia article by using the statistics of bursty periods containing the editor's edits. The integrated value of the DoI over all relevant timescales reveals which is dominant between individual-driven and interaction-driven burstiness. We empirically find that this value tends to be larger for weaker temporal correlations in the editor's editing behavior and/or stronger editorial correlations. These empirical findings are successfully confirmed by deriving an analytic form of the DoI from a model capturing the essential features of the edit sequence. Thus our approach provides a deeper insight into the origin and underlying mechanisms of bursts in human social dynamics., Comment: 12 pages, 5 figures

We have developed a much sensitve technique to conduct magnetometry under ultrahigh pressures up to 6.3~GPa, which can detect a weak volume susceptibilities as small as $\sim 10^{-4}$. An opposed-anvil-type high-pressure cell is designed by numerical analysis to give nearly zero magnetic response, in a commercial SQUID magnetometer. We introduced procedures for subtracting background contributions from a high-pressure cell by taking displacements of the cell parts into account, and found a way of resolving tiny magnetism of a sample from given magnetometer response curves. A non-magnetic material, binderless tungsten carbide ceramic, is employed. To increase sample-signal-to-background ratio further, a conical shaped gasket and cupped anvils are introduced, yielding nearly ten times better space efficiency. The new set-up and analysis are applied to measure the paramagnetic susceptibilities of spin orbit entangled moment under pressures., Comment: 9 pages, 9 figures, to appear in J. Phys. Soc. Jpn

In this study, a novel method for extracting named entities and relations from unstructured text based on the table representation is presented. By using contextualized word embeddings, the proposed method computes representations for entity mentions and long-range dependencies without complicated hand-crafted features or neural-network architectures. We also adapt a tensor dot-product to predict relation labels all at once without resorting to history-based predictions or search strategies. These advances significantly simplify the model and algorithm for the extraction of named entities and relations. Despite its simplicity, the experimental results demonstrate that the proposed method outperforms the state-of-the-art methods on the CoNLL04 and ACE05 English datasets. We also confirm that the proposed method achieves a comparable performance with the state-of-the-art NER models on the ACE05 datasets when multiple sentences are provided for context aggregation., Comment: An extended version of this paper has been accepted at Journal of Natural Language Processing

In a neutrinoless double-beta decay ($0\nu\beta\beta$) experiment, energy resolution is important to distinguish between $0\nu\beta\beta$ and background events. CAlcium fluoride for studies of Neutrino and Dark matters by Low Energy Spectrometer (CANDLES) discerns the $0\nu\beta\beta$ of $^{48}$Ca using a CaF$_2$ scintillator as the detector and source. Photomultiplier tubes (PMTs) collect scintillation photons. At the Q-value of $^{48}$Ca, the current energy resolution (2.6%) exceeds the ideal statistical fluctuation of the number of photoelectrons (1.6%). Because of CaF$_2$'s long decay constant of 1000 ns, a signal integration within 4000 ns is used to calculate the energy. The baseline fluctuation ($\sigma_{baseline}$) is accumulated in the signal integration, thus degrading the energy resolution. This paper studies $\sigma_{baseline}$ in the CANDLES detector, which severely degrades the resolution by 1% at the Q-value of $^{48}$Ca. To avoid $\sigma_{\rm baseline}$, photon counting can be used to obtain the number of photoelectrons in each PMT; however, a significant photoelectron signal overlapping probability in each PMT causes missing photoelectrons in counting and reduces the energy resolution. "Partial photon counting" reduces $\sigma_{baseline}$ and minimizes photoelectron loss. We obtain improved energy resolutions of 4.5-4.0% at 1460.8 keV ($\gamma$-ray of $^{40}$K), and 3.3-2.9% at 2614.5 keV ($\gamma$-ray of $^{208}$Tl). The energy resolution at the Q-value is estimated to be improved from 2.6% to 2.2%, and the detector sensitivity for the $0\nu\beta\beta$ half-life of $^{48}$Ca can be improved by 1.09 times., Comment: This manuscript is prepared for submission to IEEE Transaction on Nuclear Science. It contains 11 pages, 19 figures, and 1 table

We developed a CANDLES-III system to study the neutrino-less double beta (0$\nu\beta\beta$) decay of $^{48}$Ca. The proposed system employs 96 CaF$_{2}$ scintillation crystals (305 kg) with natural Ca ($^{\rm nat.}$Ca) isotope which corresponds 350\,g of $^{48}$Ca. External backgrounds were rejected using a 4$\pi$ active shield of a liquid scintillator surrounding the CaF$_2$ crystals. The internal backgrounds caused by the radioactive impurities within the CaF$_2$ crystals can be reduced effectively through analysis of the signal pulse shape. We analyzed the data obtained in the Kamioka underground for a live-time of 130.4\,days to evaluate the feasibility of the low background measurement with the CANDLES-III detector. Using Monte Carlo simulations, we estimated the background rate from the radioactive impurities in the CaF$_{2}$ crystals and the rate of high energy $\gamma$-rays caused by the (n, $\gamma$) reactions induced by environmental neutrons. The expected background rate was in a good agreement with the measured rate, i.e., approximately 10$^{-3}$ events/keV/yr/(kg of $^{\rm nat.}$Ca), in the 0$\nu\beta\beta$ window. In conclusion, the background candidates were estimated properly by comparing the measured energy spectrum with the background simulations. With this measurement method, we performed the first search for 0$\nu\beta\beta$ decay in a low background condition using a detector with a Ca isotope, in which the Ca present was not enriched, in a scale of hundreds of kg. The $^{48}$Ca isotope has a high potential for use in 0$\nu\beta\beta$ decay search, and is expected to be useful for the development of a next-generation detector for highly sensitive measurements., Comment: 26 pages, 9 figures, submitted to Physical Review D

We study distances on zigzag persistence modules from the viewpoint of derived categories and Auslander--Reiten quivers. The derived category of ordinary persistence modules is derived equivalent to that of arbitrary zigzag persistence modules, depending on a classical tilting module. Through this derived equivalence, we define and compute distances on the derived category of arbitrary zigzag persistence modules and prove an algebraic stability theorem. We also compare our distance with the distance for purely zigzag persistence modules introduced by Botnan--Lesnick and the sheaf-theoretic convolution distance due to Kashiwara--Schapira., Comment: 39 pages, 6 figures

Model-based meta-reinforcement learning (RL) methods have recently been shown to be a promising approach to improving the sample efficiency of RL in multi-task settings. However, the theoretical understanding of those methods is yet to be established, and there is currently no theoretical guarantee of their performance in a real-world environment. In this paper, we analyze the performance guarantee of model-based meta-RL methods by extending the theorems proposed by Janner et al. (2019). On the basis of our theoretical results, we propose Meta-Model-Based Meta-Policy Optimization (M3PO), a model-based meta-RL method with a performance guarantee. We demonstrate that M3PO outperforms existing meta-RL methods in continuous-control benchmarks., Comment: ACML 2021. Video demo: https://drive.google.com/file/d/1DRA-pmIWnHGNv5G_gFrml8YzKCtMcGnu/view?usp=sharing URL Source code: https://github.com/TakuyaHiraoka/Meta-Model-Based-Meta-Policy-Optimization