Search

Purpose. Photoplethysmography (PPG) is a non-invasive technique that measures changes in blood flow volume through optical means. Previous research has established the feasibility of PPG peak detection based on the crossover of moving averages. This paper proposes the Smoothed Recuarsive Moving Average Crossover, which eliminates the need for post-processing and nonlinear pre-processing of previous crossover-based peak detectors. The proposed model is advantageous regarding memory and computational complexity, making it attractive for implementations on embedded devices. Methods. Along with this paper, we make available a novel dataset comprising 66 minutes of PPG recordings. The optimization and assessment of the proposed peak detection model use this dataset. Its optimization is accomplished with the simple random search heuristic, while the leave-subject-out cross-validation method provides the means to assess its performance. The source code for all experiments reported in this research is also available in an online repository. Results. The experimental study examines the performance of the proposed model considering different arrangements of the PPG data. The experiments show that the proposed model performs better than the previous crossover-based approach from the literature regarding the precision and recall metrics. More specifically, our model has an average precision of 0.9937 and an average recall of 0.9968. Conclusion. The contribution of this research to the scientific community and literature is twofold. The dataset we collected is open for any researcher, and we improve upon the leading edge on crossover-based PPG peak detection. This improvement comes in terms of performance metrics and computational cost., Comment: 11 pages, 7 figures, 4 tables

Few-shot image classification aims to learn an image classifier using only a small set of labeled examples per class. A recent research direction for improving few-shot classifiers involves augmenting the labelled samples with synthetic images created by state-of-the-art text-to-image generation models. Following this trend, we propose Diversified In-domain Synthesis with Efficient Fine-tuning (DISEF), a novel approach which addresses the generalization challenge in few-shot learning using synthetic data. DISEF consists of two main components. First, we propose a novel text-to-image augmentation pipeline that, by leveraging the real samples and their rich semantics coming from an advanced captioning model, promotes in-domain sample diversity for better generalization. Second, we emphasize the importance of effective model fine-tuning in few-shot recognition, proposing to use Low-Rank Adaptation (LoRA) for joint adaptation of the text and image encoders in a Vision Language Model. We validate our method in ten different benchmarks, consistently outperforming baselines and establishing a new state-of-the-art for few-shot classification. Code is available at https://github.com/vturrisi/disef., Comment: 14 pages, 6 figures, 8 tables

An electrical current that flows across individual atoms or molecules can generate exotic quantum-based behavior, from memristive effects to Coulomb blockade and the promotion of quantum excited states. These fundamental effects typically appear one at a time in model junctions built using atomic tip or lateral techniques. So far, however, a viable industrial pathway for such discrete state devices has been lacking. Here, we demonstrate that a commercialized device platform can serve as this industrial pathway for quantum technologies. We have studied magnetic tunnel junctions with a MgO barrier containing C atoms. The paramagnetic localized electrons due to individual C atoms generate parallel nanotransport paths across the micronic device as deduced from magnetotransport experiments. Coulomb blockade effects linked to tunnelling magnetoresistance peaks can be electrically controlled, leading to a persistent memory effect. Our results position MgO tunneling spintronics as a promising platform to industrially implement quantum technologies.

In an effort to reduce annotation costs in action recognition, unsupervised video domain adaptation methods have been proposed that aim to adapt a predictive model from a labelled dataset (i.e., source domain) to an unlabelled dataset (i.e., target domain). In this work we address a more realistic scenario, called open-set video domain adaptation (OUVDA), where the target dataset contains "unknown" semantic categories that are not shared with the source. The challenge lies in aligning the shared classes of the two domains while separating the shared classes from the unknown ones. In this work we propose to address OUVDA with an unified contrastive learning framework that learns discriminative and well-clustered features. We also propose a video-oriented temporal contrastive loss that enables our method to better cluster the feature space by exploiting the freely available temporal information in video data. We show that discriminative feature space facilitates better separation of the unknown classes, and thereby allows us to use a simple similarity based score to identify them. We conduct thorough experimental evaluation on multiple OUVDA benchmarks and show the effectiveness of our proposed method against the prior art., Comment: Currently under review at Computer Vision and Image Understanding (CVIU) journal

Foundational image-language models have generated considerable interest due to their efficient adaptation to downstream tasks by prompt learning. Prompt learning treats part of the language model input as trainable while freezing the rest, and optimizes an Empirical Risk Minimization objective. However, Empirical Risk Minimization is known to suffer from distributional shifts which hurt generalizability to prompts unseen during training. By leveraging the regularization ability of Bayesian methods, we frame prompt learning from the Bayesian perspective and formulate it as a variational inference problem. Our approach regularizes the prompt space, reduces overfitting to the seen prompts and improves the prompt generalization on unseen prompts. Our framework is implemented by modeling the input prompt space in a probabilistic manner, as an a priori distribution which makes our proposal compatible with prompt learning approaches that are unconditional or conditional on the image. We demonstrate empirically on 15 benchmarks that Bayesian prompt learning provides an appropriate coverage of the prompt space, prevents learning spurious features, and exploits transferable invariant features. This results in better generalization of unseen prompts, even across different datasets and domains. Code available at: https://github.com/saic-fi/Bayesian-Prompt-Learning, Comment: Accepted at ICCV 2023

Over the last few years, Unsupervised Domain Adaptation (UDA) techniques have acquired remarkable importance and popularity in computer vision. However, when compared to the extensive literature available for images, the field of videos is still relatively unexplored. On the other hand, the performance of a model in action recognition is heavily affected by domain shift. In this paper, we propose a simple and novel UDA approach for video action recognition. Our approach leverages recent advances on spatio-temporal transformers to build a robust source model that better generalises to the target domain. Furthermore, our architecture learns domain invariant features thanks to the introduction of a novel alignment loss term derived from the Information Bottleneck principle. We report results on two video action recognition benchmarks for UDA, showing state-of-the-art performance on HMDB$\leftrightarrow$UCF, as well as on Kinetics$\rightarrow$NEC-Drone, which is more challenging. This demonstrates the effectiveness of our method in handling different levels of domain shift. The source code is available at https://github.com/vturrisi/UDAVT., Comment: Accepted at ICPR 2022

In the context of generative models, text-to-image generation achieved impressive results in recent years. Models using different approaches were proposed and trained in huge datasets of pairs of texts and images. However, some methods rely on pre-trained models such as Generative Adversarial Networks, searching through the latent space of the generative model by using a gradient-based approach to update the latent vector, relying on loss functions such as the cosine similarity. In this work, we follow a different direction by proposing the use of Covariance Matrix Adaptation Evolution Strategy to explore the latent space of Generative Adversarial Networks. We compare this approach to the one using Adam and a hybrid strategy. We design an experimental study to compare the three approaches using different text inputs for image generation by adapting an evaluation method based on the projection of the resulting samples into a two-dimensional grid to inspect the diversity of the distributions. The results evidence that the evolutionary method achieves more diversity in the generation of samples, exploring different regions of the resulting grids. Besides, we show that the hybrid method combines the explored areas of the gradient-based and evolutionary approaches, leveraging the quality of the results.

The nanoscopic characterization of ferroelectric thin films is crucial from their device application point of view. Standard characterization techniques are based on detecting the nanoscopic charge compensation current (switching current) caused by the polarization reversal in the ferroelectric. Owing to various surface and bulk limited mechanisms, leakage currents commonly appear during such measurements, which are frequently subtracted using the device I-V characteristic by employing positive-up-negative-down (PUND) measurement scheme. By performing nanoscopic switching current measurements on a commonly used ferroelectric, BiFeO3, we show that such characterization methods may be prone to large errors in the polarization estimation on ferro-resistive samples, due to current background subtraction issues. Especially, when ferro-resistive behavior is associated with the polarization reversal of the ferroelectric thin film, background current subtraction is not accurate due to the mismatch of the I-V characteristics for the two polarization states. We show instead that removing the background current by an asymmetric least squares subtraction method, though not perfect, gives a much better estimation of the ferroelectric properties of the sample under study.

Self-supervised models have been shown to produce comparable or better visual representations than their supervised counterparts when trained offline on unlabeled data at scale. However, their efficacy is catastrophically reduced in a Continual Learning (CL) scenario where data is presented to the model sequentially. In this paper, we show that self-supervised loss functions can be seamlessly converted into distillation mechanisms for CL by adding a predictor network that maps the current state of the representations to their past state. This enables us to devise a framework for Continual self-supervised visual representation Learning that (i) significantly improves the quality of the learned representations, (ii) is compatible with several state-of-the-art self-supervised objectives, and (iii) needs little to no hyperparameter tuning. We demonstrate the effectiveness of our approach empirically by training six popular self-supervised models in various CL settings.

Owing to prospective energy-efficient and environmentally benign applications, organic ferroelectric materials are useful and necessary alternative to inorganic ferroelectrics. Although the first discovered ferroelectric, Rochelle salt, was a salt of an organic compound, organic ferroelectrics have not been as abundant as the inorganic ones. Further, the small polarization values in the organic systems discovered so far have been a demotivating factor for their applications. However, scientific interest and activities surrounding such materials, for the purpose of fundamental understanding and practical applications, have significantly risen lately, especially after the discovery of above-room-temperature ferroelectricity in croconic acid (4,5-dihydroxy-4-cyclopentene-1,2,3-trione, H2C5O5) crystals with polarization values rivalling those found in inorganic ferroelectrics. Its large polarization, organic nature, and vacuum sublimability make croconic acid an ideal candidate for non-toxic and lead-free device applications. In this review article, we survey the scientific activities carried out so far involving ferroelectricity in this novel material, paying equal attention to its bulk single crystal and thin film forms. While we discuss about the origin of ferroelectric order and the reversal of polarization in the bulk form, we also summarize the directions toward applications of the thin films.

This paper presents solo-learn, a library of self-supervised methods for visual representation learning. Implemented in Python, using Pytorch and Pytorch lightning, the library fits both research and industry needs by featuring distributed training pipelines with mixed-precision, faster data loading via Nvidia DALI, online linear evaluation for better prototyping, and many additional training tricks. Our goal is to provide an easy-to-use library comprising a large amount of Self-supervised Learning (SSL) methods, that can be easily extended and fine-tuned by the community. solo-learn opens up avenues for exploiting large-budget SSL solutions on inexpensive smaller infrastructures and seeks to democratize SSL by making it accessible to all. The source code is available at https://github.com/vturrisi/solo-learn., Comment: Accepted to JMLR

It was discovered in 2010 that Croconic Acid, in its crystal form, has the highest polarization among organic ferroelectrics. In the context of eliminating toxic substances from electronic devices, Croconic Acid has a great potential as a sublimable lead-free ferroelectric. However, studies on ferroelectric properties of its thin films are only in their early stages and its capability to be incorporated in nanoscale devices is unknown. In this work, we demonstrate, upon ferroelectric switching at the nanoscale, stable and enduring room temperature polarization with no leakage current in Croconic Acid thin films. We thus show that it is a promising lead-free organic ferroelectric toward integration in nanoscale devices. The challenging switching current and polarization reversal characterization at the nanoscale was done using a unique combination of piezoresponse force microscopy, polarization switching current spectroscopy and the concurrent electromechanical strain response. Indeed, this combination can help to rationalize otherwise asymmetric polarization-voltage data and distorted hysteresis due to current jumps below the background noise, which are statistically washed away in macrojunctions but become prevalent at the nanoscale. These results are valid irrespective of the ferroelectrics' nature, organic or inorganic. Beyond the potential of Croconic Acid as an ecological ferroelectric material in devices, our detection of a clear nanoscopic polarization switching current thus paves the way for a fundamental understanding and technological applications of the polarization reversal mechanism at the nanoscale.

Generative Adversarial Networks (GANs) are powerful generative models that achieved strong results, mainly in the image domain. However, the training of GANs is not trivial, presenting some challenges tackled by different strategies. Evolutionary algorithms, such as COEGAN, were recently proposed as a solution to improve the GAN training, overcoming common problems that affect the model, such as vanishing gradient and mode collapse. In this work, we propose an evaluation method based on t-distributed Stochastic Neighbour Embedding (t-SNE) to assess the progress of GANs and visualize the distribution learned by generators in training. We propose the use of the feature space extracted from trained discriminators to evaluate samples produced by generators and from the input dataset. A metric based on the resulting t-SNE maps and the Jaccard index is proposed to represent the model quality. Experiments were conducted to assess the progress of GANs when trained using COEGAN. The results show both by visual inspection and metrics that the Evolutionary Algorithm gradually improves discriminators and generators through generations, avoiding problems such as mode collapse., Comment: To be published in EvoApplications 2021

Recent theory and experiments have showcased how to harness quantum mechanics to assemble heat/information engines with efficiencies that surpass the classical Carnot limit. So far, this has required atomic engines that are driven by cumbersome external electromagnetic sources. Here, using molecular spintronics, we propose an implementation that is both electronic and autonomous. Our spintronic quantum engine heuristically deploys several known quantum assets by having a chain of spin qubits formed by the paramagnetic Co centers of phthalocyanine (Pc) molecules electronically interact with electron-spin selecting Fe/C60 interfaces. Density functional calculations reveal that transport fluctuations across the interface can stabilize spin coherence on the Co paramagnetic centers, which host spin flip processes. Across vertical molecular nanodevices, we measure enduring dc current generation, output power above room temperature, two quantum thermodynamical signatures of the engine's processes, and a record 89% spin polarization of current across the Fe/C60 interface. It is crucially this electron spin selection that forces, through demonic feedback and control, charge current to flow against the built-in potential barrier. Further research into spintronic quantum engines, insight into the quantum information processes within spintronic technologies, and retooling the spintronic-based information technology chain, could help accelerate the transition to clean energy., Comment: \

Generative adversarial networks (GANs) achieved relevant advances in the field of generative algorithms, presenting high-quality results mainly in the context of images. However, GANs are hard to train, and several aspects of the model should be previously designed by hand to ensure training success. In this context, evolutionary algorithms such as COEGAN were proposed to solve the challenges in GAN training. Nevertheless, the lack of diversity and premature optimization can be found in some of these solutions. We propose in this paper the application of a quality-diversity algorithm in the evolution of GANs. The solution is based on the Novelty Search with Local Competition (NSLC) algorithm, adapting the concepts used in COEGAN to this new proposal. We compare our proposal with the original COEGAN model and with an alternative version using a global competition approach. The experimental results evidenced that our proposal increases the diversity of the discovered solutions and leverage the performance of the models found by the algorithm. Furthermore, the global competition approach was able to consistently find better models for GANs., Comment: Published in GECCO 2020

The discovery of stable room temperature ferroelectricity in Croconic Acid, an organic ferroelectric material, with polarization values on par with those found in inorganic ferroelectric materials and highest among organic ferroelectric materials, has opened up possibilities to realize myriads of nano-electronic and spintronic devices based on organic ferroelectrics. Such possibilities require an adequate understanding of the ferroelectric properties of Croconic Acid grown on surfaces that are commonly employed in device fabrication. While several macroscopic studies on relatively larger crystals of Croconic Acid have been performed, studies on thin films are only in their early stages. We have grown thin films of Croconic Acid on gold and cobalt surfaces, which are commonly used in spintronic devices as metallic electrodes, and studied the ferroelectric response of the films using ex-situ Piezoresponse Force Microscopy at room temperature. We show that the polarization reversal in Croconic Acid domains is sensitive to the substrate surface. Using the same experimental protocol, we observe the robust polarization reversal of a single, mostly in-plane electrical domain for a cobalt substrate, whereas polarization reversal is hardly observed for a gold substrate. We attribute this difference to the influence of substrates on the Croconic Acid molecular networks. Our study suggests that to realize devices one has to take care about the substrate on which Croconic Acid will be deposited. The fact that polarization switching is robust on cobalt surface can be used to fabricate multifunctional devices that utilize the cobalt-Croconic Acid interface., Comment: 21 pages, 14 figures

Generative Adversarial Networks (GANs) are an adversarial model that achieved impressive results on generative tasks. In spite of the relevant results, GANs present some challenges regarding stability, making the training usually a hit-and-miss process. To overcome these challenges, several improvements were proposed to better handle the internal characteristics of the model, such as alternative loss functions or architectural changes on the neural networks used by the generator and the discriminator. Recent works proposed the use of evolutionary algorithms on GAN training, aiming to solve these challenges and to provide an automatic way to find good models. In this context, COEGAN proposes the use of coevolution and neuroevolution to orchestrate the training of GANs. However, previous experiments detected that some of the fitness functions used to guide the evolution are not ideal. In this work we propose the evaluation of a game-based fitness function to be used within the COEGAN method. Skill rating is a metric to quantify the skill of players in a game and has already been used to evaluate GANs. We extend this idea using the skill rating in an evolutionary algorithm to train GANs. The results show that skill rating can be used as fitness to guide the evolution in COEGAN without the dependence of an external evaluator., Comment: Published in EvoApplications 2020

Generative adversarial networks (GAN) present state-of-the-art results in the generation of samples following the distribution of the input dataset. However, GANs are difficult to train, and several aspects of the model should be previously designed by hand. Neuroevolution is a well-known technique used to provide the automatic design of network architectures which was recently expanded to deep neural networks. COEGAN is a model that uses neuroevolution and coevolution in the GAN training algorithm to provide a more stable training method and the automatic design of neural network architectures. COEGAN makes use of the adversarial aspect of the GAN components to implement coevolutionary strategies in the training algorithm. Our proposal was evaluated in the Fashion-MNIST and MNIST dataset. We compare our results with a baseline based on DCGAN and also with results from a random search algorithm. We show that our method is able to discover efficient architectures in the Fashion-MNIST and MNIST datasets. The results also suggest that COEGAN can be used as a training algorithm for GANs to avoid common issues, such as the mode collapse problem., Comment: Published in GECCO 2019. arXiv admin note: text overlap with arXiv:1912.06172

Generative adversarial networks (GAN) became a hot topic, presenting impressive results in the field of computer vision. However, there are still open problems with the GAN model, such as the training stability and the hand-design of architectures. Neuroevolution is a technique that can be used to provide the automatic design of network architectures even in large search spaces as in deep neural networks. Therefore, this project proposes COEGAN, a model that combines neuroevolution and coevolution in the coordination of the GAN training algorithm. The proposal uses the adversarial characteristic between the generator and discriminator components to design an algorithm using coevolution techniques. Our proposal was evaluated in the MNIST dataset. The results suggest the improvement of the training stability and the automatic discovery of efficient network architectures for GANs. Our model also partially solves the mode collapse problem., Comment: Published in EvoApplications 2019

As more data are produced each day, and faster, data stream mining is growing in importance, making clear the need for algorithms able to fast process these data. Data stream mining algorithms are meant to be solutions to extract knowledge online, specially tailored from continuous data problem. Many of the current algorithms for data stream mining have high processing and memory costs. Often, the higher the predictive performance, the higher these costs. To increase predictive performance without largely increasing memory and time costs, this paper introduces a novel algorithm, named Online Local Boosting (OLBoost), which can be combined into online decision tree algorithms to improve their predictive performance without modifying the structure of the induced decision trees. For such, OLBoost applies a boosting to small separate regions of the instances space. Experimental results presented in this paper show that by using OLBoost the online learning decision tree algorithms can significantly improve their predictive performance. Additionally, it can make smaller trees perform as good or better than larger trees., Comment: To appear on the 8th Brazilian Conference on Intelligent Systems (BRACIS)

Dealing with memory and time constraints are current challenges when learning from data streams with a massive amount of data. Many algorithms have been proposed to handle these difficulties, among them, the Very Fast Decision Tree (VFDT) algorithm. Although the VFDT has been widely used in data stream mining, in the last years, several authors have suggested modifications to increase its performance, putting aside memory concerns by proposing memory-costly solutions. Besides, most data stream mining solutions have been centred around ensembles, which combine the memory costs of their weak learners, usually VFDTs. To reduce the memory cost, keeping the predictive performance, this study proposes the Strict VFDT (SVFDT), a novel algorithm based on the VFDT. The SVFDT algorithm minimises unnecessary tree growth, substantially reducing memory usage and keeping competitive predictive performance. Moreover, since it creates much more shallow trees than VFDT, SVFDT can achieve a shorter processing time. Experiments were carried out comparing the SVFDT with the VFDT in 11 benchmark data stream datasets. This comparison assessed the trade-off between accuracy, memory, and processing time. Statistical analysis showed that the proposed algorithm obtained similar predictive performance and significantly reduced processing time and memory use. Thus, SVFDT is a suitable option for data stream mining with memory and time limitations, recommended as a weak learner in ensemble-based solutions., Comment: 7 pages, 26 figures, Under R1 revision in Pattern Recognition Letters

We performed SQUID and FMR magnetometry experiments to clarify the relationship between two reported magnetic exchange effects arising from interfacial spin-polarized charge transfer within ferromagnetic metal (FM)/molecule bilayers: the magnetic hardening effect, and spinterface-stabilized molecular spin chains. To disentangle these effects, both of which can affect the FM magnetization reversal, we tuned the metal phthalocyanine molecule central site's magnetic moment to selectively enhance or suppress the formation of spin chains within the molecular film. We find that both effects are distinct, and additive. In the process, we 1) extended the list of FM/molecule candidate pairs that are known to generate magnetic exchange effects, 2) experimentally confirmed the predicted increase in anisotropy upon molecular adsorption; and 3) showed that spin chains within the molecular film can enhance magnetic exchange. This magnetic ordering within the organic layer implies a structural ordering. Thus, by distengangling the magnetic hardening and exchange bias contributions, our results confirm, as an echo to progress regarding inorganic spintronic tunnelling, that the milestone of spintronic tunnelling across structurally ordered organic barriers has been reached through previous magnetotransport experiments. This paves the way for solid-state devices studies that exploit the quantum physical properties of spin chains, notably through external stimuli., Comment: None

This paper presents a large amount of observations for the $\delta$ Scuti star AN Lyncis carried out in 2001--2012. The extensive observations include two tri-continent campaigns coordinated in 2002 and 2011, respectively, and several single-site contributions throughout the period. The data in total have more than 104100 raw CCD frames and photoelectric records, which consist of 165 nights (about 816 hours) spanning over 3778 days. The final reduced light curves have more than 26500 data points (including those 3462 unpublished BYU data), from which we determined 306 new timings of maximum light. A time-dependent behaviour study based on all available data indicates cyclic amplitude variability as well as period change [for the main periodicity]. Orbital sinusoid fittings to $(O-C)$ residuals and pulsation amplitudes may account for their variations being caused by the light-time effect of AN Lyn in a binary system. The orbital period is about 26--30 years. Current results support the binarity of AN Lyn, first suspected by Zhou (2002). We further show the detailed time evolution structure of the pulsation of AN Lyn as function of both time and period through wavelet analyses., Comment: 23 figures, 5 tables, reduced data, maxima, and raw data available. Submitted initial manuscript, in revision in response to referee's report