Search

Sound content is an indispensable element for multimedia works such as video games, music, and films. Recent high-quality diffusion-based sound generation models can serve as valuable tools for the creators. However, despite producing high-quality sounds, these models often suffer from slow inference speeds. This drawback burdens creators, who typically refine their sounds through trial and error to align them with their artistic intentions. To address this issue, we introduce Sound Consistency Trajectory Models (SoundCTM). Our model enables flexible transitioning between high-quality 1-step sound generation and superior sound quality through multi-step generation. This allows creators to initially control sounds with 1-step samples before refining them through multi-step generation. While CTM fundamentally achieves flexible 1-step and multi-step generation, its impressive performance heavily depends on an additional pretrained feature extractor and an adversarial loss, which are expensive to train and not always available in other domains. Thus, we reframe CTM's training framework and introduce a novel feature distance by utilizing the teacher's network for a distillation loss. Additionally, while distilling classifier-free guided trajectories, we train conditional and unconditional student models simultaneously and interpolate between these models during inference. We also propose training-free controllable frameworks for SoundCTM, leveraging its flexible sampling capability. SoundCTM achieves both promising 1-step and multi-step real-time sound generation without using any extra off-the-shelf networks. Furthermore, we demonstrate SoundCTM's capability of controllable sound generation in a training-free manner. Our codes, pretrained models, and audio samples are available at https://github.com/sony/soundctm., Comment: Audio samples: https://koichi-saito-sony.github.io/soundctm/. Codes: https://github.com/sony/soundctm. Checkpoints: https://huggingface.co/Sony/soundctm

Techniques that explain the predictions of black-box machine learning models are crucial to make the models transparent, thereby increasing trust in AI systems. The input features to the models often have a nested structure that consists of high- and low-level features, and each high-level feature is decomposed into multiple low-level features. For such inputs, both high-level feature attributions (HiFAs) and low-level feature attributions (LoFAs) are important for better understanding the model's decision. In this paper, we propose a model-agnostic local explanation method that effectively exploits the nested structure of the input to estimate the two-level feature attributions simultaneously. A key idea of the proposed method is to introduce the consistency property that should exist between the HiFAs and LoFAs, thereby bridging the separate optimization problems for estimating them. Thanks to this consistency property, the proposed method can produce HiFAs and LoFAs that are both faithful to the black-box models and consistent with each other, using a smaller number of queries to the models. In experiments on image classification in multiple instance learning and text classification using language models, we demonstrate that the HiFAs and LoFAs estimated by the proposed method are accurate, faithful to the behaviors of the black-box models, and provide consistent explanations.

The main objective of this paper is to Identify which macroe conomic factors and industrial indexes influenced the total Brazilian banking spread between March 2011 and March 2015. This paper considers subclassification of industrial activities in Brazil. Monthly time series data were used in multivariate linear regression models using Eviews (7.0). Eighteen variables were considered as candidates to be determinants. Variables which positively influenced bank spread are; Default, IPIs (Industrial Production Indexes) for capital goods, intermediate goods, du rable consumer goods, semi-durable and non-durable goods, the Selic, GDP, unemployment rate and EMBI +. Variables which influence negatively are; Consumer and general consumer goods IPIs, IPCA, the balance of the loan portfolio and the retail sales index. A p-value of 05% was considered. The main conclusion of this work is that the progress of industry, job creation and consumption can reduce bank spread. Keywords: Credit. Bank spread. Macroeconomics. Industrial Production Indexes. Finance., Comment: in Portuguese langauge. Keywords: Credit. Bank spread. Macroeconomics. Industrial Production Indexes. Finance

In several real-world applications in medical and control engineering, there are unsafe solutions whose evaluations involve inherent risk. This optimization setting is known as safe optimization and formulated as a specialized type of constrained optimization problem with constraints for safety functions. Safe optimization requires performing efficient optimization without evaluating unsafe solutions. A few studies have proposed the optimization methods for safe optimization based on Bayesian optimization and the evolutionary algorithm. However, Bayesian optimization-based methods often struggle to achieve superior solutions, and the evolutionary algorithm-based method fails to effectively reduce unsafe evaluations. This study focuses on CMA-ES as an efficient evolutionary algorithm and proposes an optimization method termed safe CMA-ES. The safe CMA-ES is designed to achieve both safety and efficiency in safe optimization. The safe CMA-ES estimates the Lipschitz constants of safety functions transformed with the distribution parameters using the maximum norm of the gradient in Gaussian process regression. Subsequently, the safe CMA-ES projects the samples to the nearest point in the safe region constructed with the estimated Lipschitz constants. The numerical simulation using the benchmark functions shows that the safe CMA-ES successfully performs optimization, suppressing the unsafe evaluations, while the existing methods struggle to significantly reduce the unsafe evaluations., Comment: This paper has been accepted as a full paper at GECCO2024

Black-box optimization problems often require simultaneously optimizing different types of variables, such as continuous, integer, and categorical variables. Unlike integer variables, categorical variables do not necessarily have a meaningful order, and the discretization approach of continuous variables does not work well. Although several Bayesian optimization methods can deal with mixed-category black-box optimization (MC-BBO), they suffer from a lack of scalability to high-dimensional problems and internal computational cost. This paper proposes CatCMA, a stochastic optimization method for MC-BBO problems, which employs the joint probability distribution of multivariate Gaussian and categorical distributions as the search distribution. CatCMA updates the parameters of the joint probability distribution in the natural gradient direction. CatCMA also incorporates the acceleration techniques used in the covariance matrix adaptation evolution strategy (CMA-ES) and the stochastic natural gradient method, such as step-size adaptation and learning rate adaptation. In addition, we restrict the ranges of the categorical distribution parameters by margin to prevent premature convergence and analytically derive a promising margin setting. Numerical experiments show that the performance of CatCMA is superior and more robust to problem dimensions compared to state-of-the-art Bayesian optimization algorithms., Comment: This paper has been accepted for presentation at GECCO2024. We have modified the equations in Table 1

Evolving consumer demands and market trends have led to businesses increasingly embracing a production approach that prioritizes flexibility and customization. Consequently, factory workers must engage in tasks that are more complex than before. Thus, productivity depends on each worker's skills in assembling products. Therefore, analyzing the behavior of a worker is crucial for work improvement. However, manual analysis is time consuming and does not provide quick and accurate feedback. Machine learning have been attempted to automate the analyses; however, most of these methods need several labels for training. To this end, we extend the Gaussian process hidden semi-Markov model (GP-HSMM), to enable the rapid and automated analysis of worker behavior without pre-training. The model does not require labeled data and can automatically and accurately segment continuous motions into motion classes. The proposed model is a probabilistic model that hierarchically connects GP-HSMM and HSMM, enabling the extraction of behavioral patterns with different granularities. Furthermore, it mutually infers the parameters between the GP-HSMM and HSMM, resulting in accurate motion pattern extraction. We applied the proposed method to motion data in which workers assembled products at an actual production site. The accuracy of behavior pattern extraction was evaluated using normalized Levenshtein distance (NLD). The smaller the value of NLD, the more accurate is the pattern extraction. The NLD of motion patterns captured by GP-HSMM and HSMM layers in our proposed method was 0.50 and 0.33, respectively, which are the smallest compared to that of the baseline methods.

The HNC/HCN ratio is observationally known as a thermometer in Galactic interstellar molecular clouds. A recent study has alternatively suggested that the HNC/HCN ratio is affected by the ultraviolet (UV) field, not by the temperature. We aim to study this ratio on the scale of giant molecular clouds in the barred spiral galaxy M83 towards the southwestern bar end and the central region from ALMA observations, and if possible, distinguish the above scenarios. We compare the high (40-50 pc) resolution HNC/HCN ratios with the star formation rate from the 3-mm continuum intensity and the molecular mass inferred from the HCN intensities. Our results show that the HNC/HCN ratios do not vary with the star formation rates, star formation efficiencies, or column densities in the bar-end region. In the central region, the HNC/HCN ratios become higher with higher star formation rates, which tend to cause higher temperatures. This result is not consistent with the previously proposed scenario in which the HNC/HCN ratio decreases with increasing temperature. Spectral shapes suggest that this trend may be due to optically thick HCN and optically thin HNC. In addition, we compare the large-scale ($\sim 200$ pc) correlation between the dust temperature from the FIR ratio and the HNC/HCN ratio for the southwestern bar-end region. The HNC/HCN ratio is lower when the dust temperatures are higher. We suggest from the above results that the HNC/HCN ratio depends on the UV radiation field that affects the interstellar medium on the $\sim100\,$pc scale where the column densities are low., Comment: 17 pages, 9 figures. Accepted for publication in ApJ

When a liquid is rapidly cooled below its melting point without inducing crystallization, its dynamics slow down significantly without noticeable structural changes. Elucidating the origin of this slowdown has been a long-standing challenge. Here, we report a theoretical investigation into the mechanism of the dynamic slowdown in supercooled water, a ubiquitous yet extraordinary substance characterized by various anomalous properties arising from local density fluctuations. Using molecular dynamics simulations, we found that the jump dynamics, which are elementary structural change processes, deviate from Poisson statistics with decreasing temperature. This deviation is attributed to slow variables competing with the jump motions, i.e., dynamic disorder. The present analysis of the dynamic disorder showed that the primary slow variable is the displacement of the fourth nearest oxygen atom of a jumping molecule, which occurs in an environment created by the fluctuations of molecules outside the first hydration shell. As the temperature decreases, the jump dynamics become slow and intermittent. These intermittent dynamics are attributed to the prolonged trapping of jumping molecules within extended and stable low-density domains. As the temperature continues to decrease, the number of slow variables increases due to the increased cooperative motions. Consequently, the jump dynamics proceed in a higher-dimensional space consisting of multiple slow variables, becoming slower and more intermittent. It is then conceivable that with further decreasing temperature, the slowing and intermittency of the jump dynamics intensify, eventually culminating in a glass transition.

Quantum search algorithms, such as Grover's algorithm, are expected to efficiently solve constrained combinatorial optimization problems. However, implementing a quantum search algorithm for solving the traveling salesman problem (TSP) on a circuit poses a potential challenge because current quantum search algorithms for TSP assume that an initial state of equal superposition of feasible solution states satisfying the constraint is already prepared a priori. The time complexity of brute-force preparation of the initial state increases exponentially with the factorial growth of feasible solutions, posing a considerable obstacle in designing quantum circuits for large-scale TSP. To overcome this problem, we propose a two-step quantum search algorithm with two distinct operators for preparing the initial state and solving TSP. The algorithm first amplifies an equal superposition state of all feasible solutions of TSP and subsequently amplifies the optimal solution states among these feasible solution states. Our algorithm, encoded in the higher-order unconstrained binary optimization (HOBO) representation, notably reduces the required number of qubits, enabling efficient preparation of the initial state with a unified circuit design and solving TSP with a quadratic speedup in the absence of prior knowledge of feasible solutions., Comment: submitted to 2024 IEEE International Conference on Quantum Computing and Engineering (QCE24)

This study introduces an uncertainty-aware, mesh-free numerical method for solving Kolmogorov PDEs. In the proposed method, we use Gaussian process regression (GPR) to smoothly interpolate pointwise solutions that are obtained by Monte Carlo methods based on the Feynman-Kac formula. The proposed method has two main advantages: 1. uncertainty assessment, which is facilitated by the probabilistic nature of GPR, and 2. mesh-free computation, which allows efficient handling of high-dimensional PDEs. The quality of the solution is improved by adjusting the kernel function and incorporating noise information from the Monte Carlo samples into the GPR noise model. The performance of the method is rigorously analyzed based on a theoretical lower bound on the posterior variance, which serves as a measure of the error between the numerical and true solutions. Extensive tests on three representative PDEs demonstrate the high accuracy and robustness of the method compared to existing methods., Comment: 17 pages, 3 figures

We report low-temperature powder X-ray and neutron diffraction studies on breathing pyrochlore magnets Cu$M$Cr$_{4}$S$_{8}$ ($M$ = Al, Ga), which undergo a magnetic transition at $T_{\rm N} \approx$ 21 and 31 K for {$M$ = Al and Ga, respectively. X-ray diffraction reveals that the magnetic transition accompanies a structural transition from cubic $F{\overline 4}3m$ to polar orthorhombic $Imm2$ symmetry for both the compounds, with larger distortion observed for $M$ = Ga at low temperatures. Neutron diffraction reveals incommensurate magnetic modulation ${\mathbf Q} = (q_{\rm IC}, 0.5, 0)$ in the orthorhombic setting, where $q_{\rm IC} \approx$ 0.39 and 0.31 for $M$ = Al and Ga, respectively. Our magnetic-structure analysis suggests cycloid-type magnetic order but not proper-screw type for both the compounds. We find strong correlation between the local spin configuration and Cr-Cr bond lengths, indicating that the spin-lattice coupling as well as the magnetic frustration play an important role in determining the ground state. Cu$M$Cr$_{4}$S$_{8}$ potentially offers a platform to explore magnetoelectric effects arising from the helimagnet driven electric polarity., Comment: 8 pages, 6 figures, 5 tables

The Quantum Alternating Operator Ansatz (QAOA) represents a branch of quantum algorithms for solving combinatorial optimization problems. A specific variant, the Grover-Mixer Quantum Alternating Operator Ansatz (GM-QAOA), ensures uniform amplitude across states that share equivalent objective values. This property makes the algorithm independent of the problem structure, focusing instead on the distribution of objective values within the problem. In this work, we prove the probability upper bound for measuring a computational basis state from a GM-QAOA circuit with a given depth, which is a critical factor in QAOA cost. Using this, we derive the upper bounds for the probability of sampling an optimal solution, and for the approximation ratio of maximum optimization problems, both dependent on the objective value distribution. Through numerical analysis, we link the distribution to the problem size and build the regression models that relate the problem size, QAOA depth, and performance upper bound. Our results suggest that the GM-QAOA provides a quadratic enhancement in sampling probability and requires circuit depth that scales exponentially with problem size to maintain consistent performance., Comment: 19 pages, 7 figures, 1 table

Probabilistic Error Cancellation (PEC) aims to improve the accuracy of expectation values for observables.This is accomplished using the probabilistic insertion of recovery gates, which correspond to the inverse of errors.However, the inserted recovery gates also induce errors. Thus, it is difficult to obtain accurate expectation values with PEC since the estimator of PEC has a bias due to noise induced by recovery gates.To address this challenge, we propose an improved version of PEC that considers the noise resulting from gate insertion, called Feed-Forward PEC (FFPEC). FFPEC provides an unbiased estimator of expectation values by cancelling out the noise induced by recovery gates.We demonstrate that FFPEC yields more accurate expectation values compared to conventional PEC method through numerical simulations with bit-flip and depolarizing noises.

Variational Quantum Algorithms (VQAs) represent a class of algorithms that utilize a hybrid approach, combining classical and quantum computing techniques. In this approach, classical computers serve as optimizers that update circuit parameters to find approximate solutions to complex problems. In this study, we apply a method known as Light Cone Cancellation (LCC) to optimize variational circuits, effectively reducing the required number of qubits and gates for circuit simulation. We then evaluate the performance of LCC one of the VQAs -- the Variational Quantum Eigensolver (VQE) -- to address the Max-Cut problem. Compared with the Quantum Approximate Optimization Algorithm (QAOA), VQE offers greater degrees of freedom at lower circuit depths. By applying LCC to VQE, we can shift the complexity of circuit simulation from the number of qubits to the number of edges in the graph, i.e., from exponential time to polynomial time. This enables us to solve large problems up to 50 vertices, without actually simulating the entire circuit. From our simulation in a 7-qubit and a 27-qubit noisy devices, we show that LCC yields higher approximation ratios than those cases without LCC, implying that the effect of noise is reduced when LCC is applied.

Let $\lfloor x\rfloor$ denote the integer part of $x$. In 1947, Mills constructed a real number $\xi$ greater than $1$ such that $\lfloor \xi^{3^k} \rfloor$ is always a prime number for every positive integer $k$. We define Mills' constant as the smallest real number $\xi$ satisfying this property. In this article, we determine that Mills' constant is irrational. Moreover, we also obtain partial results on the transcendency of the constant., Comment: 11 pages

Discriminantal arrangements are hyperplane arrangements, which are generalized braid ones. They are constructed from given hyperplane arrangements, but their combinatorics are not invariant under combinatorial equivalence. However, it is known that the combinatorics of the discriminantal arrangement are constant on a Zariski open set of the space of hyperplane arrangements. In the present paper, we introduce non-very generic varieties in the space of hyperplane arrangements to classify discriminantal arrangements and show that the Zariski open set is the complement of non-very generic varieties. We study their basic properties and construction and provide examples, including infinite families of non-very generic varieties. In particular, the construction we call degeneration is a powerful tool for constructing non-very generic varieties. As an application, we provide lists of non-very generic varieties for spaces of small line arrangements., Comment: 14 pages. There were mistakes in the statement of Proposition 4.3 and 4.4 in the old version, so they have been corrected

Recent advancements in self-supervised learning in the point cloud domain have demonstrated significant potential. However, these methods often suffer from drawbacks, including lengthy pre-training time, the necessity of reconstruction in the input space, or the necessity of additional modalities. In order to address these issues, we introduce Point-JEPA, a joint embedding predictive architecture designed specifically for point cloud data. To this end, we introduce a sequencer that orders point cloud tokens to efficiently compute and utilize tokens proximity based on their indices during target and context selection. The sequencer also allows shared computations of the tokens proximity between context and target selection, further improving the efficiency. Experimentally, our method achieves competitive results with state-of-the-art methods while avoiding the reconstruction in the input space or additional modality., Comment: 6 pages, 3 figures

Short- and long-term outcomes of an algorithm often differ, with damaging downstream effects. A known example is a click-bait algorithm, which may increase short-term clicks but damage long-term user engagement. A possible solution to estimate the long-term outcome is to run an online experiment or A/B test for the potential algorithms, but it takes months or even longer to observe the long-term outcomes of interest, making the algorithm selection process unacceptably slow. This work thus studies the problem of feasibly yet accurately estimating the long-term outcome of an algorithm using only historical and short-term experiment data. Existing approaches to this problem either need a restrictive assumption about the short-term outcomes called surrogacy or cannot effectively use short-term outcomes, which is inefficient. Therefore, we propose a new framework called Long-term Off-Policy Evaluation (LOPE), which is based on reward function decomposition. LOPE works under a more relaxed assumption than surrogacy and effectively leverages short-term rewards to substantially reduce the variance. Synthetic experiments show that LOPE outperforms existing approaches particularly when surrogacy is severely violated and the long-term reward is noisy. In addition, real-world experiments on large-scale A/B test data collected on a music streaming platform show that LOPE can estimate the long-term outcome of actual algorithms more accurately than existing feasible methods., Comment: TheWebConference 2024

There has been a growing interest in off-policy evaluation in the literature such as recommender systems and personalized medicine. We have so far seen significant progress in developing estimators aimed at accurately estimating the effectiveness of counterfactual policies based on biased logged data. However, there are many cases where those estimators are used not only to evaluate the value of decision making policies but also to search for the best hyperparameters from a large candidate space. This work explores the latter hyperparameter optimization (HPO) task for off-policy learning. We empirically show that naively applying an unbiased estimator of the generalization performance as a surrogate objective in HPO can cause an unexpected failure, merely pursuing hyperparameters whose generalization performance is greatly overestimated. We then propose simple and computationally efficient corrections to the typical HPO procedure to deal with the aforementioned issues simultaneously. Empirical investigations demonstrate the effectiveness of our proposed HPO algorithm in situations where the typical procedure fails severely., Comment: IJCAI'24

We investigate the ground state and dynamics of a mixture of spin-1 and spin-2 Bose-Einstein condensates of ${}^{87}{\rm{Rb}}$ atoms. For the experimentally measured interaction coefficients, the ground state exhibits phase separation between the spin-1 ferromagnetic state and the spin-2 nematic state. At the interface between them, a partially polarized spin state emerges. The uniformly mixed state of the spin-1 polar state and spin-2 biaxial nematic state is metastable, and the phase separation via nucleation can be triggered by a local perturbation., Comment: 9 pages, 5 figures, 3 movies

It is known that the set of all nonnegative integers may be equipped with a total order that is chaotic in the sense that there is no monotone three-term arithmetic progressions. Such chaotic order must be so complicated that the resulting ordered set cannot be order isomorphic to the set of all nonnegative integers or the set of all integers with the standard order. In this paper, we completely characterize order structures of chaotic orders on the set of all nonnegative integers, as well as on the set of all integers and on the set of all rational numbers., Comment: 8 pages

This study investigates the escape of Mercury's sodium-group ions (Na+-group, including ions with m/q from 21 to 30 amu/e) and their dependence on true anomaly angle (TAA), i.e., Mercury's orbital phase around the Sun, using measurements from MESSENGER. The measurements are categorized into solar wind, magnetosheath, and magnetosphere, and further divided into four TAA intervals. Na+-group ions form escape plumes in the solar wind and magnetosheath, with higher fluxes along the solar wind's motional electric field. The total escape rates vary from 0.2 to 1 times 10^{25} atoms/s with the magnetosheath being the main escaping region. These rates exhibit a TAA dependence, peaking near the perihelion and similar during Mercury's remaining orbit. Despite Mercury's tenuous exosphere, Na+-group ions escape rate is comparable to other inner planets. This can be attributed to several processes, including that Na+-group ions may include several ion species, efficient photoionization frequency for elements within Na+-group, etc., Comment: 21 pages, 4 figures

We investigate the quantum spin-1/2 zigzag chain with frustrated $J_1$-$J_2$ Heisenberg interactions, incorporating additional off-diagonal exchange interactions known as the $\Gamma$ term, both with and without an applied magnetic field. Based on the density-matrix renormalization group calculation, we map out the ground state phase diagram that shows a variety of magnetic and nonmagnetic phases including multicritical points and several exactly solvable points. Upon introducing a finite $\Gamma$ term, we observe the persistent dimer singlet state of the $J_1$-$J_2$ Heisenberg model, sustaining a nonzero spin gap, while also giving rise to a gapless nonmagnetic excitation, manifesting in the substantial zero-energy peak in the nematic dynamical structure factor. This gapless peak-mode remaining almost as a fluctuation to the ground state, induces dilute but robust concentration of nematicity on top of singlets on dimers, which we call the nematic singlet-dimer phase. When the whole nematic excited mode condenses and replaces the singlet, the nematic-dimer phase transforms to the Ising-type ferromagnetic or antiferromagnetic long-range orders that arise from the $\Gamma$ term spontaneously selecting magnetic easy axes. Its orientations dictate the type of magnetic order under geometric frustration effects as predicted by Landau's mean-field theory. These theoretical findings provide insights into the exotic low-temperature phase observed in YbCuS$_2$, characterized by gapless excitations and seemingly nonmagnetic behavior accompanied by incommensurate correlations.

Preceding a core-collapse supernova, various processes produce an increasing amount of neutrinos of all flavors characterized by mounting energies from the interior of massive stars. Among them, the electron antineutrinos are potentially detectable by terrestrial neutrino experiments such as KamLAND and Super-Kamiokande via inverse beta decay interactions. Once these pre-supernova neutrinos are observed, an early warning of the upcoming core-collapse supernova can be provided. In light of this, KamLAND and Super-Kamiokande, both located in the Kamioka mine in Japan, have been monitoring pre-supernova neutrinos since 2015 and 2021, respectively. Recently, we performed a joint study between KamLAND and Super-Kamiokande on pre-supernova neutrino detection. A pre-supernova alert system combining the KamLAND detector and the Super-Kamiokande detector was developed and put into operation, which can provide a supernova alert to the astrophysics community. Fully leveraging the complementary properties of these two detectors, the combined alert is expected to resolve a pre-supernova neutrino signal from a 15 M$_{\odot}$ star within 510 pc of the Earth, at a significance level corresponding to a false alarm rate of no more than 1 per century. For a Betelgeuse-like model with optimistic parameters, it can provide early warnings up to 12 hours in advance., Comment: Resubmitted to ApJ. 22 pages, 16 figures, for more information about the combined pre-supernova alert system, see https://www.lowbg.org/presnalarm/

We have selected six myths about the OSS community and have tested whether they are true or not. The purpose of this report is to identify the lessons that can be learned from the development style of the OSS community and the issues that need to be addressed in order to achieve better Employee Experience (EX) in software development within companies and organizations. The OSS community has been led by a group of skilled developers known as hackers. We have great respect for the engineers and activities of the OSS community and aim to learn from them. On the other hand, it is important to recognize that having high expectations can sometimes result in misunderstandings. When there are excessive expectations and concerns, misunderstandings (referred to as myths) can arise, particularly when individuals who are not practitioners rely on hearsay to understand the practices of practitioners. We selected the myths to be tested based on a literature review and interviews. These myths are held by software development managers and customers who are not direct participants in the OSS community. We answered questions about each myth through: 1) Our own analysis of repository data, 2) A literature survey of data analysis conducted by previous studies, or 3) A combination of the two approaches., Comment: English Version: 28 pages + Japanese Version: 23 pages

Observations with distributed sensors are essential in analyzing a series of human and machine activities (referred to as 'events' in this paper) in complex and extensive real-world environments. This is because the information obtained from a single sensor is often missing or fragmented in such an environment; observations from multiple locations and modalities should be integrated to analyze events comprehensively. However, a learning method has yet to be established to extract joint representations that effectively combine such distributed observations. Therefore, we propose Guided Masked sELf-Distillation modeling (Guided-MELD) for inter-sensor relationship modeling. The basic idea of Guided-MELD is to learn to supplement the information from the masked sensor with information from other sensors needed to detect the event. Guided-MELD is expected to enable the system to effectively distill the fragmented or redundant target event information obtained by the sensors without being overly dependent on any specific sensors. To validate the effectiveness of the proposed method in novel tasks of distributed multimedia sensor event analysis, we recorded two new datasets that fit the problem setting: MM-Store and MM-Office. These datasets consist of human activities in a convenience store and an office, recorded using distributed cameras and microphones. Experimental results on these datasets show that the proposed Guided-MELD improves event tagging and detection performance and outperforms conventional inter-sensor relationship modeling methods. Furthermore, the proposed method performed robustly even when sensors were reduced., Comment: 13page, 7figure, under review

We investigate the four-dimensional Wess-Zumino-Witten (WZW) terms within the framework of $Sp$ quantum chromodynamics (QCD) using invertible field theory through bordism theory. We present a novel approach aimed at circumventing both perturbative and non-perturbative gauge anomalies on spacetime manifolds endowed with spin structures. We study both ungauged and gauged WZW terms including the problems of the topological consistency of gauged WZW terms., Comment: 20 pages

We present Eagle (RWKV-5) and Finch (RWKV-6), sequence models improving upon the RWKV (RWKV-4) architecture. Our architectural design advancements include multi-headed matrix-valued states and a dynamic recurrence mechanism that improve expressivity while maintaining the inference efficiency characteristics of RNNs. We introduce a new multilingual corpus with 1.12 trillion tokens and a fast tokenizer based on greedy matching for enhanced multilinguality. We trained four Eagle models, ranging from 0.46 to 7.5 billion parameters, and two Finch models with 1.6 and 3.1 billion parameters and find that they achieve competitive performance across a wide variety of benchmarks. We release all our models on HuggingFace under the Apache 2.0 license. Models at: https://huggingface.co/RWKV Training code at: https://github.com/RWKV/RWKV-LM Inference code at: https://github.com/RWKV/ChatRWKV Time-parallel training code at: https://github.com/RWKV/RWKV-infctx-trainer

We present observations of the central 9 kpc of the Antennae merger (NGC 4038/9) at 55 pc resolution in the CO 2-1 line obtained with the Atacama Large Millimeter/submillimeter Array (ALMA). We use a pixel-based analysis to compare the gas properties in the Antennae to those in 70 nearby spiral galaxies from the PHANGS-ALMA survey, as well as the merger and nearest luminous infrared galaxy NGC 3256. Compared to PHANGS galaxies at matched spatial resolution, the molecular gas in the Antennae exhibits some of the highest surface densities, velocity dispersions, peak brightness temperatures, and turbulent pressures. However, the virial parameters in the Antennae are consistent with many of the PHANGS galaxies. NGC 3256 has similar gas surface densities but higher nuclear velocity dispersions than the Antennae, as well as higher system-wide peak brightness temperatures and virial parameters. NGC 3256 is at a later stage in the merging process than the Antennae, which may result in more intense merger-driven gas flows that could drive up the turbulence in the gas. The high virial parameters in NGC 3256 may indicate that this increased turbulence is suppressing future star formation as NGC 3256 moves out of the starburst phase. In comparison, the relatively normal virial parameters in the Antennae may imply that it is about to undergo a new burst of star formation., Comment: 16 pages, 8 figures, accepted to MNRAS

The area law of the bipartite information measure characterizes one of the most fundamental aspects of quantum many-body physics. In thermal equilibrium, the area law for the mutual information universally holds at arbitrary temperatures as long as the systems have short-range interactions. In systems with power-law decaying interactions, $r^{-\alpha}$ ($r$: distance), conditions for the thermal area law are elusive. In this work, we aim to clarify the optimal condition $\alpha> \alpha_c$ such that the thermal area law universally holds. A standard approach to considering the conditions is to focus on the magnitude of the boundary interaction between two subsystems. However, we find here that the thermal area law is more robust than this conventional argument suggests. We show the optimal threshold for the thermal area law by $\alpha_c= (D+1)/2$ ($D$: the spatial dimension of the lattice), assuming a power-law decay of the clustering for the bipartite correlations. Remarkably, this condition encompasses even the thermodynamically unstable regimes $\alpha < D$. We verify this condition numerically, finding that it is qualitatively accurate for both integrable and non-integrable systems. Unconditional proof of the thermal area law is possible by developing the power-law clustering theorem for $\alpha > D$ above a threshold temperature. Furthermore, the numerical calculation for the logarithmic negativity shows that the same criterion $\alpha > (D+1)/2$ applies to the thermal area law for quantum entanglement., Comment: 16 pages, 6 figures

H-poor superluminous supernovae (SLSNe-I) are characterized by O II lines around 4,000 - 4,500 A in pre-/near-maximum spectra, so-called W-shaped O II lines. As these lines are from relatively high excitation levels, they have been considered a sign of non-thermal processes, which may give a hint of power sources of SLSNe-I. However, the conditions for these lines to appear have not been understood well. In this work, we systematically calculate synthetic spectra to reproduce observed spectra of eight SLSNe-I, parameterizing departure coefficients from the nebular approximation in the SN ejecta (expressed as b_neb). We find that most of the observed spectra can be reproduced well with b_neb ~< 10, which means that no strong departure is necessary for the formation of the W-shaped O II lines. We also show that the appearance of the W-shaped O II lines is sensitive to temperature; only spectra with temperatures T ~ 14,000 - 16,000 K can produce the W-shaped O II lines without large departures. Based on this, we constrain the non-thermal ionization rate near the photosphere. Our results suggest that spectral features of SLSNe-I can give independent constraints on the power source through the non-thermal ionization rates., Comment: 19 pages, 11 figures, accepted for publication in ApJ

We present a quantitative and statistical analysis of the molecular gas morphology in 73 nearby galaxies using high spatial resolution CO (J=2-1) data obtained from the Atacama Large Millimeter/submillimeter Array (ALMA) by the PHANGS large program. We applied three model-independent parameters: Concentration (C), Asymmetry (A), and Clumpiness (S) which are commonly used to parameterize the optical and infrared morphology of galaxies. We find a significant correlation between A and S, with a Spearman's rank correlation coefficient of 0.62 with a p-value of $4 \times 10^{-9}$. This suggests a higher abundance of molecular clumps (i.e. giant molecular cloud associations) in galaxies that display stronger distortion or biased large-scale molecular gas distribution. In addition, the analysis of the C parameter suggests high central molecular concentration in most barred spiral galaxies investigated in this study. Furthermore, we found a positive correlation between the length of the bar structure ($R_\mathrm{bar}/R_{25}$) and the C parameter, with a Spearman's rank correlation coefficient of 0.63 with a p-value of $3.8 \times 10^{-5}$ suggesting that larger bar structure can facilitate overall molecular gas transport and yield higher central concentration than galaxies with shorter bars. Finally, we offer a possible classification scheme of nearby disk galaxies which is based on the CAS parameters of molecular gas., Comment: 23 pages, 18 figures, 4 Tables

We present InterHandGen, a novel framework that learns the generative prior of two-hand interaction. Sampling from our model yields plausible and diverse two-hand shapes in close interaction with or without an object. Our prior can be incorporated into any optimization or learning methods to reduce ambiguity in an ill-posed setup. Our key observation is that directly modeling the joint distribution of multiple instances imposes high learning complexity due to its combinatorial nature. Thus, we propose to decompose the modeling of joint distribution into the modeling of factored unconditional and conditional single instance distribution. In particular, we introduce a diffusion model that learns the single-hand distribution unconditional and conditional to another hand via conditioning dropout. For sampling, we combine anti-penetration and classifier-free guidance to enable plausible generation. Furthermore, we establish the rigorous evaluation protocol of two-hand synthesis, where our method significantly outperforms baseline generative models in terms of plausibility and diversity. We also demonstrate that our diffusion prior can boost the performance of two-hand reconstruction from monocular in-the-wild images, achieving new state-of-the-art accuracy., Comment: Accepted to CVPR 2024, project page: https://jyunlee.github.io/projects/interhandgen/

Conventional machine learning algorithms have traditionally been designed under the assumption that input data follows a vector-based format, with an emphasis on vector-centric paradigms. However, as the demand for tasks involving set-based inputs has grown, there has been a paradigm shift in the research community towards addressing these challenges. In recent years, the emergence of neural network architectures such as Deep Sets and Transformers has presented a significant advancement in the treatment of set-based data. These architectures are specifically engineered to naturally accommodate sets as input, enabling more effective representation and processing of set structures. Consequently, there has been a surge of research endeavors dedicated to exploring and harnessing the capabilities of these architectures for various tasks involving the approximation of set functions. This comprehensive survey aims to provide an overview of the diverse problem settings and ongoing research efforts pertaining to neural networks that approximate set functions. By delving into the intricacies of these approaches and elucidating the associated challenges, the survey aims to equip readers with a comprehensive understanding of the field. Through this comprehensive perspective, we hope that researchers can gain valuable insights into the potential applications, inherent limitations, and future directions of set-based neural networks. Indeed, from this survey we gain two insights: i) Deep Sets and its variants can be generalized by differences in the aggregation function, and ii) the behavior of Deep Sets is sensitive to the choice of the aggregation function. From these observations, we show that Deep Sets, one of the well-known permutation-invariant neural networks, can be generalized in the sense of a quasi-arithmetic mean.

We investigate the control mechanism of trap frequencies and determine the orientation of ion trap principal axes. The application of DC voltage to the ground electrodes, commonly employed to finely tune trap frequencies in ion traps, leads to the rotation of the trap principal axes. Analyzing the ion matter-wave interference signal enables us to determine the directions of the trap principal axes. Both the experiments and simulations reveal an avoided-crossing behavior resulting from the coupling between the trap radial axes. Additionally, simulations indicate that symmetric trap structures lack this coupling, suggesting that trap structure asymmetry causes coupling between the axes. The findings of this study offer valuable insights into ion traps for diverse applications in quantum science and technology., Comment: 6 pages, 3 figures

We present UTDUSS, the UTokyo-SaruLab system submitted to Interspeech2024 Speech Processing Using Discrete Speech Unit Challenge. The challenge focuses on using discrete speech unit learned from large speech corpora for some tasks. We submitted our UTDUSS system to two text-to-speech tracks: Vocoder and Acoustic+Vocoder. Our system incorporates neural audio codec (NAC) pre-trained on only speech corpora, which makes the learned codec represent rich acoustic features that are necessary for high-fidelity speech reconstruction. For the acoustic+vocoder track, we trained an acoustic model based on Transformer encoder-decoder that predicted the pre-trained NAC tokens from text input. We describe our strategies to build these models, such as data selection, downsampling, and hyper-parameter tuning. Our system ranked in second and first for the Vocoder and Acoustic+Vocoder tracks, respectively., Comment: 5 pages, 3 figures