Search

Accurate representation of the multiscale features in spatiotemporal physical systems using vision transformer (ViT) architectures requires extremely long, computationally prohibitive token sequences. To address this issue, we propose two adaptive tokenization schemes that dynamically adjust patch sizes based on local features: one ensures convergent behavior to uniform patch refinement, while the other offers better computational efficiency. Moreover, we present a set of spatiotemporal attention schemes, where the temporal or axial spatial dimensions are decoupled, and evaluate their computational and data efficiencies. We assess the performance of the proposed multiscale adaptive model, MATEY, in a sequence of experiments. The results show that adaptive tokenization schemes achieve improved accuracy without significantly increasing the length of the token sequence. Compared to a full spatiotemporal attention scheme or a scheme that decouples only the temporal dimension, we find that fully decoupled axial attention is less efficient and expressive, requiring more training time and model weights to achieve the same accuracy. Finally, we demonstrate in two fine-tuning tasks featuring different physics that models pretrained on PDEBench data outperform the ones trained from scratch, especially in the low data regime with frozen attention.

We show that in the asymptotic large-volume limit, the original proposal for Quantum Spectral Curve for AdS3 x S3 x T4 with R-R flux has a wider class of solutions, than studied previously. We argue that in this limit the QSC reduces to a finite set of Bethe equations for both massive and massless particle types. We also find that the QSC imposes more constraining conditions on the dressing phases than previously known and we present solutions of those equations.

Superconducting circuits are being employed for large-scale quantum devices, and a pertinent challenge is to perform accurate numerical simulations of device parameters. One of the most advanced methods for analyzing superconducting circuit designs is the energy participation ratio (EPR) method, which constructs quantum Hamiltonians based on the energy distribution extracted from classical electromagnetic simulations. In the EPR approach, we extract linear terms from finite element simulations and add nonlinear terms using the energy participation ratio extracted from the classical simulations. However, the EPR method relies on a low-order expansion of nonlinear terms, which is prohibitive for accurately describing highly anharmonic circuits. An example of such a circuit is the fluxonium qubit, which has recently attracted increasing attention due to its high lifetimes and low error rates. In this work, we extend the EPR approach to effectively address highly nonlinear superconducting circuits, and, as a proof of concept, we apply our approach to a fluxonium qubit. Specifically, we design, fabricate, and experimentally measure a fluxonium qubit coupled to a readout resonator. We compare the measured frequencies of both the qubit and the resonator to those extracted from the EPR analysis, and we find an excellent agreement. Furthermore, we compare the dispersive shift as a function of external flux obtained from experiments with our EPR analysis and a simpler lumped element model. Our findings reveal that the EPR results closely align with the experimental data, providing more accurate estimations compared to the simplified lumped element simulations., Comment: 13 pages, 8 figures

The ability to perform rapid, high fidelity readout of a qubit state is an important requirement for quantum algorithms and, in particular, for enabling operations such as mid-circuit measurements and measurement-based feedback for error correction schemes on large quantum processors. The growing interest in fluxonium qubits, due to their long coherence times and high anharmonicity, merits further attention to reducing the readout duration and measurement errors. We find that this can be accomplished by exploiting the flux tunability of fluxonium qubits. In this work, we experimentally demonstrate flux-pulse-assisted readout, as proposed in Phys. Rev. Applied 22, 014079 (https://doi.org/10.1103/PhysRevApplied.22.014079), in a setup without a quantum-limited parametric amplifier. Increasing the dispersive shift magnitude by almost 20% through flux pulsing, we achieve an assignment fidelity of 94.3% with an integration time of 280 ns. The readout performance is limited by state initialization, but we find that the limit imposed only by the signal-to-noise ratio corresponds to an assignment fidelity of 99.9% with a 360 ns integration time. We also verify these results through simple semi-classical simulations. These results constitute the fastest reported readout of a fluxonium qubit, with the prospect of further improvement by incorporation of a parametric amplifier in the readout chain to enhance measurement efficiency., Comment: 9 pages, 6 figures, 1 table

Symbolic music research plays a crucial role in music-related machine learning, but MIDI data can be complex for those without musical expertise. To address this issue, we present MidiTok Visualizer, a web application designed to facilitate the exploration and visualization of various MIDI tokenization methods from the MidiTok Python package. MidiTok Visualizer offers numerous customizable parameters, enabling users to upload MIDI files to visualize tokenized data alongside an interactive piano roll., Comment: in Extended Abstracts for the Late-Breaking Demo Sessionof the 25th Int. Society for Music Information Retrieval Conf., San Francisco, United States, 2024

Consumer electronics used to follow the miniaturization trend described by Moore's Law. Despite increased processing power in Microcontroller Units (MCUs), MCUs used in the smallest appliances are still not capable of running even moderately big, state-of-the-art artificial neural networks (ANNs) especially in time-sensitive scenarios. In this work, we present a novel method called Scattered Online Inference (SOI) that aims to reduce the computational complexity of ANNs. SOI leverages the continuity and seasonality of time-series data and model predictions, enabling extrapolation for processing speed improvements, particularly in deeper layers. By applying compression, SOI generates more general inner partial states of ANN, allowing skipping full model recalculation at each inference., Comment: NeurIPS 2024

This paper proposes a definition of recognizable transducers over monads and comonads, which bridges two important ongoing efforts in the current research on regularity. The first effort is the study of regular transductions, which extends the notion of regularity from languages into word-to-word functions. The other important effort is generalizing the notion of regular languages from words to arbitrary monads, introduced in arXiv:1502.04898. In this paper, we present a number of examples of transducer classes that fit the proposed framework. In particular we show that our class generalizes the classes of Mealy machines and rational transductions. We also present examples of recognizable transducers for infinite words and a specific type of trees called terms. The main result of this paper is a theorem, which states the class of recognizable transductions is closed under composition, subject to some coherence axioms between the structure of a monad and the structure of a comonad. Due to its complexity, we formalize the proof of the theorem in Coq Proof Assistant. In the proof, we introduce the concepts of a context and a generalized wreath product for Eilenberg-Moore algebras, which could be valuable tools for studying these algebras., Comment: 33 pages

This thesis studies the single-use restriction for register automata and transducers over infinite alphabets. The restriction requires that a read-access to a register should have the side effect of destroying its contents. This constraint results in robust classes of languages and transductions. For automata models, we show that one-way register automata, two-way register automata, and orbit-finite monoids have the same expressive power. For transducer models, we show that single-use Mealy machines and single-use two-way transducers admit versions of the Krohn-Rhodes decomposition theorem. Moreover, single-use Mealy machines are equivalent to an algebraic model called local algebraic semigroup transductions. Additionally, we show that single-use two-way transducers are equivalent to single-use streaming string transducers (SSTs) over infinite alphabets and to regular list functions with atoms. Compared with the previous work arXiv:1907.10504, this thesis offers a coherent narrative on the single-use restriction. We introduce an abstract notion of single-use functions and use them to define all the discussed single-use models. We also introduce and study the algebraic models of local semigroup transduction and local rational semigroup transduction., Comment: PhD Thesis at University of Warsaw. Supervisor: Miko{\l}aj Boja\'nczyk

Data structures and algorithms are essential building blocks for programs, and \emph{distributed data structures}, which automatically partition data across multiple memory locales, are essential to writing high-level parallel programs. While many projects have designed and implemented C++ distributed data structures and algorithms, there has not been widespread adoption of an interoperable model allowing algorithms and data structures from different libraries to work together. This paper introduces distributed ranges, which is a model for building generic data structures, views, and algorithms. A distributed range extends a C++ range, which is an iterable sequence of values, with a concept of segmentation, thus exposing how the distributed range is partitioned over multiple memory locales. Distributed data structures provide this distributed range interface, which allows them to be used with a collection of generic algorithms implemented using the distributed range interface. The modular nature of the model allows for the straightforward implementation of \textit{distributed views}, which are lightweight objects that provide a lazily evaluated view of another range. Views can be composed together recursively and combined with algorithms to implement computational kernels using efficient, flexible, and high-level standard C++ primitives. We evaluate the distributed ranges model by implementing a set of standard concepts and views as well as two execution runtimes, a multi-node, MPI-based runtime and a single-process, multi-GPU runtime. We demonstrate that high-level algorithms implemented using generic, high-level distributed ranges can achieve performance competitive with highly-tuned, expert-written code., Comment: To appear in ACM International Conference on Supercomputing (ICS) 2024

Orbit-finite sets are a generalisation of finite sets, and as such support many operations allowed for finite sets, such as pairing, quotienting, or taking subsets. However, they do not support function spaces, i.e. if X and Y are orbit-finite sets, then the space of finitely supported functions from X to Y is not orbit-finite. In this paper we propose two solutions to this problem: one is obtained by generalising the notion of orbit-finite set, and the other one is obtained by restricting it. In both cases, function spaces and the original closure properties are retained. Curiously, both solutions are "linear": the generalisation is based on linear algebra, while the restriction is based on linear logic.

String theory on AdS${}_3\times$ S${}^3\times$ T${}^4$ geometries supported by a combination of NS-NS and R-R charges is believed to be integrable. We elucidate the kinematics and analytic structure of worldsheet excitations in mixed charge and pure NS-NS backgrounds, when expressed in momentum, Zhukovsky variables and the rapidity $u$ which appears in the quantum spectral curve. We discuss the relations between fundamental and bound state excitations and the role of fusion in constraining and determining the S matrices of these theories. We propose a scalar dressing factor consistent with a novel $u$-plane periodicity and comment on its close relation to the XXZ model at roots of unity. We solve the odd part of crossing and show that our solution is consistent with fusion and reduces in the relativistic limit to dressing phases previously found in the literature., Comment: 58 pages, 77 figures. v2: added missing reference, fixed typos. v3: clarified discussion of physical region, added references

Much attention has focused on the transmon architecture for large-scale superconducting quantum devices, however, the fluxonium qubit has emerged as a possible successor. With a shunting inductor in parallel to a Josephson junction, the fluxonium offers larger anharmonicity and stronger protection against dielectric loss, leading to higher coherence times as compared to conventional transmon qubits. The interplay between the inductive and Josephson energy potentials of the fluxonium qubit leads to a rich dispersive shift landscape when tuning the external flux. Here we propose to exploit the features in the dispersive shift to improve qubit readout. Specifically, we report on theoretical simulations showing improved readout times and error rates by performing the readout at a flux bias point with large dispersive shift. We expand the scheme to include different error channels, and show that with an integration time of 155 ns, flux-pulse-assisted readout offers about 5 times improvement in the signal to noise ratio. Moreover, we show that the performance improvement persists in the presence of finite measurement efficiency combined with quasi-static flux noise, and also when considering the increased Purcell rate at the flux-pulse-assisted readout point. We suggest a set of reasonable energy parameters for the fluxonium architecture that will allow for the implementation of our proposed flux-pulse-assisted readout scheme., Comment: 11 pages, 7 figures

A tunneling junction between normal electrode and a topological superconducting wire, mediated by a quantum dot, is considered theoretically. We show that the presence of the dot in the junction can be advantageous to Majorana zero modes identification. Namely, we demonstrate that for the dot strongly coupled to the wire, the Majorana mode from the upper chiral sub-band "leaks" into the dot, providing supplementary information on Majorana mode formation. Thus, both the Zeeman-split dot sub-levels detect Majorana partners of a Kramers pair, formed at the wire end. The characteristic three-peak structures in both spin sectors of the spectral density of the dot, distinguish from the trivial scenario of one Andreev resonance at Fermi energy produced exclusively by the dot's spin sub-levels., Comment: paper presented at The European Conference Physics of Magnetism 2023, Pozna\'n, Poland, accepted for J. of Magnetism and Magnetic Materials

Translational research requires data at multiple scales of biological organization. Advancements in sequencing and multi-omics technologies have increased the availability of these data, but researchers face significant integration challenges. Knowledge graphs (KGs) are used to model complex phenomena, and methods exist to construct them automatically. However, tackling complex biomedical integration problems requires flexibility in the way knowledge is modeled. Moreover, existing KG construction methods provide robust tooling at the cost of fixed or limited choices among knowledge representation models. PheKnowLator (Phenotype Knowledge Translator) is a semantic ecosystem for automating the FAIR (Findable, Accessible, Interoperable, and Reusable) construction of ontologically grounded KGs with fully customizable knowledge representation. The ecosystem includes KG construction resources (e.g., data preparation APIs), analysis tools (e.g., SPARQL endpoints and abstraction algorithms), and benchmarks (e.g., prebuilt KGs and embeddings). We evaluated the ecosystem by systematically comparing it to existing open-source KG construction methods and by analyzing its computational performance when used to construct 12 large-scale KGs. With flexible knowledge representation, PheKnowLator enables fully customizable KGs without compromising performance or usability.

This paper describes the design and implementation of Spinoza, a fast and flexible quantum simulator written in Rust. Spinoza simulates the evolution of a quantum system's state by applying quantum gates, with the core design principle being that a single-qubit gate applied to a target qubit preserves the probability of pairs of amplitudes corresponding to measurement outcomes that differ only in the target qubit. Multiple strategies are employed for selecting pairs of amplitudes, depending on the gate type and other parameters, to optimize performance. Specific optimizations are also implemented for certain gate types and target qubits. Spinoza is intended to enable the development of quantum computing solutions by offering developers a simple, flexible, and fast tool for classical simulation. In this paper we provide details about the design and usage examples. Furthermore, we compare Spinoza's performance against several other open-source simulators to demonstrate its strengths., Comment: 12 pages, 9 figures

The real-time processing of time series signals is a critical issue for many real-life applications. The idea of real-time processing is especially important in audio domain as the human perception of sound is sensitive to any kind of disturbance in perceived signals, especially the lag between auditory and visual modalities. The rise of deep learning (DL) models complicated the landscape of signal processing. Although they often have superior quality compared to standard DSP methods, this advantage is diminished by higher latency. In this work we propose novel method for minimization of inference time latency and memory consumption, called Short-Term Memory Convolution (STMC) and its transposed counterpart. The main advantage of STMC is the low latency comparable to long short-term memory (LSTM) networks. Furthermore, the training of STMC-based models is faster and more stable as the method is based solely on convolutional neural networks (CNNs). In this study we demonstrate an application of this solution to a U-Net model for a speech separation task and GhostNet model in acoustic scene classification (ASC) task. In case of speech separation we achieved a 5-fold reduction in inference time and a 2-fold reduction in latency without affecting the output quality. The inference time for ASC task was up to 4 times faster while preserving the original accuracy., Comment: ICLR 2023

In this article, we apply the theory of quantum channels and open-system state evolution to propose a unitary parameterization and an efficient learning algorithm for Quantum Hidden Markov Models (QHMMs). By leveraging the richer dynamics of quantum channels, we demonstrate the greater efficiency of quantum stochastic generators compared to classical ones. Specifically, we prove that a stochastic process can be simulated within a quantum Hilbert space using quadratically fewer dimensions than in a classical stochastic vector space. We show that any QHMM can be efficiently implemented and simulated using a quantum circuit with mid-circuit measurements. A key advantage for feasible QHMM learning in the hypothesis space of unitary circuits lies in the continuity of Stinespring's dilation. Specifically, if the unitary parameterizations of channels are close in the operator norm, the corresponding channels will be close in both diamond norm and Bures distance. This property forms the foundation for defining of efficient learning algorithms with continuous fitness landscapes. By employing the unitary parameterization of QHMMs, we establish a formal generative learning model. This model formalizes the empirical distributions of target stochastic process languages, defines the hypothesis space of quantum circuits, and introduces an empirical stochastic divergence measure-hypothesis fitness-as a criterion for learning success. The smooth mapping between the hypothesis and fitness spaces facilitates the development of efficient heuristic and gradient descent algorithms. We consider four examples of stochastic process languages and train QHMMs with hyperparameter-adaptive evolutionary search and multi-parameter nonlinear optimization technique applied to parameterized quantum ansatz circuits. We confirm our results by running optimal circuits on quantum hardware., Comment: 35 pages, 28 figures

Recent developments in quantum computing and machine learning have propelled the interdisciplinary study of quantum machine learning. Sequential modeling is an important task with high scientific and commercial value. Existing VQC or QNN-based methods require significant computational resources to perform the gradient-based optimization of a larger number of quantum circuit parameters. The major drawback is that such quantum gradient calculation requires a large amount of circuit evaluation, posing challenges in current near-term quantum hardware and simulation software. In this work, we approach sequential modeling by applying a reservoir computing (RC) framework to quantum recurrent neural networks (QRNN-RC) that are based on classical RNN, LSTM and GRU. The main idea to this RC approach is that the QRNN with randomly initialized weights is treated as a dynamical system and only the final classical linear layer is trained. Our numerical simulations show that the QRNN-RC can reach results comparable to fully trained QRNN models for several function approximation and time series prediction tasks. Since the QRNN training complexity is significantly reduced, the proposed model trains notably faster. In this work we also compare to corresponding classical RNN-based RC implementations and show that the quantum version learns faster by requiring fewer training epochs in most cases. Our results demonstrate a new possibility to utilize quantum neural network for sequential modeling with greater quantum hardware efficiency, an important design consideration for noisy intermediate-scale quantum (NISQ) computers.

Background: Common data models solve many challenges of standardizing electronic health record (EHR) data, but are unable to semantically integrate all the resources needed for deep phenotyping. Open Biological and Biomedical Ontology (OBO) Foundry ontologies provide computable representations of biological knowledge and enable the integration of heterogeneous data. However, mapping EHR data to OBO ontologies requires significant manual curation and domain expertise. Objective: We introduce OMOP2OBO, an algorithm for mapping Observational Medical Outcomes Partnership (OMOP) vocabularies to OBO ontologies. Results: Using OMOP2OBO, we produced mappings for 92,367 conditions, 8611 drug ingredients, and 10,673 measurement results, which covered 68-99% of concepts used in clinical practice when examined across 24 hospitals. When used to phenotype rare disease patients, the mappings helped systematically identify undiagnosed patients who might benefit from genetic testing. Conclusions: By aligning OMOP vocabularies to OBO ontologies our algorithm presents new opportunities to advance EHR-based deep phenotyping., Comment: Supplementary Material is included at the end of the manuscript

In this paper, construction of a neural-network based, closed-loop control of a discontinuous capsule drive is analyzed. The foundation of the designed controller is an optimized open-loop control function. A neural network is used to determine the dependence between the open-loop controller's output and the system's state. Robustness of the neural controller with respect to variation of parameters of the controlled system is analyzed and compared with the original, optimized open-loop control. It is expected that the presented method can facilitate construction of closed-loop controllers of systems, for which other methods are not effective, such as non-smooth or discontinuous ones., Comment: 23 pages, 7 figures, submitted to Meccanica, Special Issue on Self-Propelled Robots: from Theory to Applications