Rodrigo Fernandes de Mello

Rodrigo Mello, Filipe Calegario, Geber Ramalho

Despite recent advancements, the field of text-to-image synthesis still suffers from lack of fine-grained control. Using only text, it remains challenging to deal with issues such as concept coherence and concept contamination. We propose a method to enhance control by generating specific concepts that can be reused throughout multiple images, effectively expanding natural language with new words that can be combined much like a painter's palette. Unlike previous contributions, our method does not copy visuals from input data and can generate concepts through text alone. We perform a set of comparisons that finds our method to be a significant improvement over text-only prompts.

Heitor Murilo Gomes, Maciej Grzenda, Rodrigo Mello et al.

Unlabelled data appear in many domains and are particularly relevant to streaming applications, where even though data is abundant, labelled data is rare. To address the learning problems associated with such data, one can ignore the unlabelled data and focus only on the labelled data (supervised learning); use the labelled data and attempt to leverage the unlabelled data (semi-supervised learning); or assume some labels will be available on request (active learning). The first approach is the simplest, yet the amount of labelled data available will limit the predictive performance. The second relies on finding and exploiting the underlying characteristics of the data distribution. The third depends on an external agent to provide the required labels in a timely fashion. This survey pays special attention to methods that leverage unlabelled data in a semi-supervised setting. We also discuss the delayed labelling issue, which impacts both fully supervised and semi-supervised methods. We propose a unified problem setting, discuss the learning guarantees and existing methods, explain the differences between related problem settings. Finally, we review the current benchmarking practices and propose adaptations to enhance them.

Gabriel D. Cantareira, Rodrigo F. Mello, Fernando V. Paulovich

As neural networks become the tool of choice to solve an increasing variety of problems in our society, adversarial attacks become critical. The possibility of generating data instances deliberately designed to fool a network's analysis can have disastrous consequences. Recent work has shown that commonly used methods for model training often result in fragile abstract representations that are particularly vulnerable to such attacks. This paper presents a visual framework to investigate neural network models subjected to adversarial examples, revealing how models' perception of the adversarial data differs from regular data instances and their relationships with class perception. Through different use cases, we show how observing these elements can quickly pinpoint exploited areas in a model, allowing further study of vulnerable features in input data and serving as a guide to improving model training and architecture.

Lucas Pagliosa, Rodrigo Mello

Concept Drift (CD) detection intends to continuously identify changes in data stream behaviors, supporting researchers in the study and modeling of real-world phenomena. Motivated by the lack of learning guarantees in current CD algorithms, we decided to take advantage of the Statistical Learning Theory (SLT) to formalize the necessary requirements to ensure probabilistic learning bounds, so drifts would refer to actual changes in data rather than by chance. As discussed along this paper, a set of mathematical assumptions must be held in order to rely on SLT bounds, which are especially controversial in CD scenarios. Based on this issue, we propose a methodology to address those assumptions in CD scenarios and therefore ensure learning guarantees. Complementary, we assessed a set of relevant and known CD algorithms from the literature in light of our methodology. As main contribution, we expect this work to support researchers while designing and evaluating CD algorithms on different domains.

Lucas Pagliosa, Alexandru Telea, Rodrigo Mello

The reconstruction of phase spaces is an essential step to analyze time series according to Dynamical System concepts. A regression performed on such spaces unveils the relationships among system states from which we can derive their generating rules, that is, the most probable set of functions responsible for generating observations along time. In this sense, most approaches rely on Takens' embedding theorem to unfold the phase space, which requires the embedding dimension and the time delay. Moreover, although several methods have been proposed to empirically estimate those parameters, they still face limitations due to their lack of consistency and robustness, which has motivated this paper. As an alternative, we here propose an artificial neural network with a forgetting mechanism to implicitly learn the phase spaces properties, whatever they are. Such network trains on forecasting errors and, after converging, its architecture is used to estimate the embedding parameters. Experimental results confirm that our approach is either as competitive as or better than most state-of-the-art strategies while revealing the temporal relationship among time-series observations.

Yule Vaz, Rodrigo Fernandes de Mello, Carlos Henrique Grossi

The Data Clustering (DC) problem is of central importance for the area of Machine Learning (ML), given its usefulness to represent data structural similarities from input spaces. Differently from Supervised Machine Learning (SML), which relies on the theoretical frameworks of the Statistical Learning Theory (SLT) and the Algorithm Stability (AS), DC has scarce literature on general-purpose learning guarantees, affecting conclusive remarks on how those algorithms should be designed as well as on the validity of their results. In this context, this manuscript introduces a new concept, based on multidimensional persistent homology, to analyze the conditions on which a clustering model is capable of generalizing data. As a first step, we propose a more general definition of DC problem by relying on Topological Spaces, instead of metric ones as typically approached in the literature. From that, we show that the DC problem presents an analogous dilemma to the Bias-Variance one, which is here referred to as the Coarse-Refinement (CR) dilemma. CR is intended to clarify the contrast between: (i) highly-refined partitions and the clustering instability (overfitting); and (ii) over-coarse partitions and the lack of representativeness (underfitting); consequently, the CR dilemma suggests the need of a relaxation of Kleinberg's richness axiom. Experimental results were used to illustrate that multidimensional persistent homology support the measurement of divergences among DC models, leading to a consistency criterion.

Rodrigo Fernandes de Mello

The Statistical Learning Theory (SLT) provides the foundation to ensure that a supervised algorithm generalizes the mapping $f: \mathcal{X} \to \mathcal{Y}$ given $f$ is selected from its search space bias $\mathcal{F}$. SLT depends on the Shattering coefficient function $\mathcal{N}(\mathcal{F},n)$ to upper bound the empirical risk minimization principle, from which one can estimate the necessary training sample size to ensure the probabilistic learning convergence and, most importantly, the characterization of the capacity of $\mathcal{F}$, including its underfitting and overfitting abilities while addressing specific target problems. However, the analytical solution of the Shattering coefficient is still an open problem since the first studies by Vapnik and Chervonenkis in $1962$, which we address on specific datasets, in this paper, by employing equivalence relations from Topology, data separability results by Har-Peled and Jones, and combinatorics. Our approach computes the Shattering coefficient for both binary and multi-class datasets, leading to the following additional contributions: (i) the estimation of the required number of hyperplanes in the worst and best-case classification scenarios and the respective $Ω$ and $O$ complexities; (ii) the estimation of the training sample sizes required to ensure supervised learning; and (iii) the comparison of dataset embeddings, once they (re)organize samples into some new space configuration. All results introduced and discussed along this paper are supported by the R package shattering (https://cran.r-project.org/web/packages/shattering).

Tiago S. Nazare, Rodrigo F. de Mello, Moacir A. Ponti

Recently, several techniques have been explored to detect unusual behaviour in surveillance videos. Nevertheless, few studies leverage features from pre-trained CNNs and none of then present a comparison of features generate by different models. Motivated by this gap, we compare features extracted by four state-of-the-art image classification networks as a way of describing patches from security video frames. We carry out experiments on the Ped1 and Ped2 datasets and analyze the usage of different feature normalization techniques. Our results indicate that choosing the appropriate normalization is crucial to improve the anomaly detection performance when working with CNN features. Also, in the Ped2 dataset our approach was able to obtain results comparable to the ones of several state-of-the-art methods. Lastly, as our method only considers the appearance of each frame, we believe that it can be combined with approaches that focus on motion patterns to further improve performance.

Rodrigo Fernandes de Mello, Moacir Antonelli Ponti, Carlos Henrique Grossi Ferreira

The Statistical Learning Theory (SLT) provides the theoretical guarantees for supervised machine learning based on the Empirical Risk Minimization Principle (ERMP). Such principle defines an upper bound to ensure the uniform convergence of the empirical risk Remp(f), i.e., the error measured on a given data sample, to the expected value of risk R(f) (a.k.a. actual risk), which depends on the Joint Probability Distribution P(X x Y) mapping input examples x in X to class labels y in Y. The uniform convergence is only ensured when the Shattering coefficient N(F,2n) has a polynomial growing behavior. This paper proves the Shattering coefficient for any Hilbert space H containing the input space X and discusses its effects in terms of learning guarantees for supervised machine algorithms.

Rodrigo Fernandes de Mello, Martha Dais Ferreira, Moacir Antonelli Ponti

Deep Learning (DL) is one of the most common subjects when Machine Learning and Data Science approaches are considered. There are clearly two movements related to DL: the first aggregates researchers in quest to outperform other algorithms from literature, trying to win contests by considering often small decreases in the empirical risk; and the second investigates overfitting evidences, questioning the learning capabilities of DL classifiers. Motivated by such opposed points of view, this paper employs the Statistical Learning Theory (SLT) to study the convergence of Deep Neural Networks, with particular interest in Convolutional Neural Networks. In order to draw theoretical conclusions, we propose an approach to estimate the Shattering coefficient of those classification algorithms, providing a lower bound for the complexity of their space of admissible functions, a.k.a. algorithm bias. Based on such estimator, we generalize the complexity of network biases, and, next, we study AlexNet and VGG16 architectures in the point of view of their Shattering coefficients, and number of training examples required to provide theoretical learning guarantees. From our theoretical formulation, we show the conditions which Deep Neural Networks learn as well as point out another issue: DL benchmarks may be strictly driven by empirical risks, disregarding the complexity of algorithms biases.

Christopher Dane Shulby, Martha Dais Ferreira, Rodrigo F. de Mello et al.

High-accuracy speech recognition is especially challenging when large datasets are not available. It is possible to bridge this gap with careful and knowledge-driven parsing combined with the biologically inspired CNN and the learning guarantees of the Vapnik Chervonenkis (VC) theory. This work presents a Shallow-CNN-HTSVM (Hierarchical Tree Support Vector Machine classifier) architecture which uses a predefined knowledge-based set of rules with statistical machine learning techniques. Here we show that gross errors present even in state-of-the-art systems can be avoided and that an accurate acoustic model can be built in a hierarchical fashion. The CNN-HTSVM acoustic model outperforms traditional GMM-HMM models and the HTSVM structure outperforms a MLP multi-class classifier. More importantly we isolate the performance of the acoustic model and provide results on both the frame and phoneme level considering the true robustness of the model. We show that even with a small amount of data accurate and robust recognition rates can be obtained.