arXiv subjects

Tiago Fonseca

Publications and source records attributed to Tiago Fonseca.

16 recordsLinked to original sources

arXiv2026

Fine-Tuning Whisper for Automatic Speech Recognition in Baniwa: A Preliminary Study

Automatic Speech Recognition (ASR) technologies have achieved remarkable performance in recent years through the use of large multilingual foundation models. However, most advances remain concentrated on high-resource languages, while indigenous languages continue to suffer from a lack of speech resources and language technologies. This work presents a preliminary study on the adaptation of Whisper for Automatic Speech Recognition in Baniwa, an indigenous Arawakan language spoken in Brazil, Colombia, and Venezuela. The experiments were conducted using a corpus of 1,373 manually transcribed recordings obtained from a linguistic documentation project. The corpus contains approximately 0.54 hours of speech and consists primarily of isolated words and short elicited utterances. The Whisper Small model was fine-tuned using supervised learning and evaluated using Word Error Rate (WER) and Character Error Rate (CER). The best model achieved a WER of 37.5% and a CER of 7.45%, demonstrating that multilingual foundation models can be successfully adapted to extremely low-resource indigenous languages. The results establish an initial baseline for Baniwa Automatic Speech Recognition and provide a foundation for future research involving larger datasets, language-specific adaptation strategies, and post-processing techniques.

cs.CL↗

arXiv2025

Percepta: High Performance Stream Processing at the Edge

The rise of real-time data and the proliferation of Internet of Things (IoT) devices have highlighted the limitations of cloud-centric solutions, particularly regarding latency, bandwidth, and privacy. These challenges have driven the growth of Edge Computing. Associated with IoT appears a set of other problems, like: data rate harmonization between multiple sources, protocol conversion, handling the loss of data and the integration with Artificial Intelligence (AI) models. This paper presents Percepta, a lightweight Data Stream Processing (DSP) system tailored to support AI workloads at the edge, with a particular focus on such as Reinforcement Learning (RL). It introduces specialized features such as reward function computation, data storage for model retraining, and real-time data preparation to support continuous decision-making. Additional functionalities include data normalization, harmonization across heterogeneous protocols and sampling rates, and robust handling of missing or incomplete data, making it well suited for the challenges of edge-based AI deployment.

cs.DC↗

arXiv2025

Control of Renewable Energy Communities using AI and Real-World Data

The electrification of transportation and the increased adoption of decentralized renewable energy generation have added complexity to managing Renewable Energy Communities (RECs). Integrating Electric Vehicle (EV) charging with building energy systems like heating, ventilation, air conditioning (HVAC), photovoltaic (PV) generation, and battery storage presents significant opportunities but also practical challenges. Reinforcement learning (RL), particularly MultiAgent Deep Deterministic Policy Gradient (MADDPG) algorithms, have shown promising results in simulation, outperforming heuristic control strategies. However, translating these successes into real-world deployments faces substantial challenges, including incomplete and noisy data, integration of heterogeneous subsystems, synchronization issues, unpredictable occupant behavior, and missing critical EV state-of-charge (SoC) information. This paper introduces a framework designed explicitly to handle these complexities and bridge the simulation to-reality gap. The framework incorporates EnergAIze, a MADDPG-based multi-agent control strategy, and specifically addresses challenges related to real-world data collection, system integration, and user behavior modeling. Preliminary results collected from a real-world operational REC with four residential buildings demonstrate the practical feasibility of our approach, achieving an average 9% reduction in daily peak demand and a 5% decrease in energy costs through optimized load scheduling and EV charging behaviors. These outcomes underscore the framework's effectiveness, advancing the practical deployment of intelligent energy management solutions in RECs.

eess.SY↗

arXiv2024

FlexiGen: Stochastic Dataset Generator for Electric Vehicle Charging Energy Flexibility

Electric vehicles (EVs) and renewable energy sources (RES) are vital components of sustainable energy systems, yet their uncoordinated integration can pose substantial challenges to grid stability, such as unmanaged peak loads and energy balance issues. Vehicle-to-Grid (V2G), offer a promising solution to address these challenges by enabling bidirectional energy flow between EVs and the grid. As such, EVs can be used in advances Demand Response (DR) strategies to optimize energy use and mitigate the intermittency of renewable generation. To reach such advantages, optimization algorithms need data on EV energy flexibility, such as charging patterns and usage preferences. However, data collection remains constrained by challenges such as high costs, user engagement, data privacy concerns, and limited access to open-source datasets on EV energy flexibility. This paper presents FlexiGen an open-source stochastic dataset generator tool designed to overcome the data limitations in EV flexibility for V2G and V1G DR applications. FlexiGen generates synthetic datasets encompassing realist EV usage patterns, behaviours and flexibility scenarios for household and office routines. To generate these datasets, FlexiGen uses a series of configurable probabilistic variables, such as stochastic user routines, traffic conditions, charger types, car average electricity consumption and State of Charge (SoC). The generated datasets include an hourly routine with the EV State of connection, Destination Charger, Estimated Departure Time, Required SOC at Departure, Estimated Arrival Time, and Estimated SOC at Arrival. Accompanying this publication an example dataset is generated for 3 households with 1 EV each, and 1 office building with 3 EVs. The generated dataset is analyzed and discussed on the paper and published alongside the open-source code for FlexiGen tool.

eess.SY↗

arXiv2024

CityLearn v2: Energy-flexible, resilient, occupant-centric, and carbon-aware management of grid-interactive communities

As more distributed energy resources become part of the demand-side infrastructure, it is important to quantify the energy flexibility they provide on a community scale, particularly to understand the impact of geographic, climatic, and occupant behavioral differences on their effectiveness, as well as identify the best control strategies to accelerate their real-world adoption. CityLearn provides an environment for benchmarking simple and advanced distributed energy resource control algorithms including rule-based, model-predictive, and reinforcement learning control. CityLearn v2 presented here extends CityLearn v1 by providing a simulation environment that leverages the End-Use Load Profiles for the U.S. Building Stock dataset to create virtual grid-interactive communities for resilient, multi-agent distributed energy resources and objective control with dynamic occupant feedback. This work details the v2 environment design and provides application examples that utilize reinforcement learning to manage battery energy storage system charging/discharging cycles, vehicle-to-grid control, and thermal comfort during heat pump power modulation.

cs.LG↗

arXiv2024

EVLearn: Extending the CityLearn Framework with Electric Vehicle Simulation

Intelligent energy management strategies, such as Vehicle-to-Grid (V2G) and Grid-to-Vehicle (G2V) emerge as a potential solution to the Electric Vehicles' (EVs) integration into the energy grid. These strategies promise enhanced grid resilience and economic benefits for both vehicle owners and grid operators. Despite the announced prospective, the adoption of these strategies is still hindered by an array of operational problems. Key among these is the lack of a simulation platform that allows to validate and refine V2G and G2V strategies. Including the development, training, and testing in the context of Energy Communities (ECs) incorporating multiple flexible energy assets. Addressing this gap, first we introduce the EVLearn, a simulation module for researching in both V2G and G2V energy management strategies, that models EVs, their charging infrastructure and associated energy flexibility dynamics; second, this paper integrates EVLearn with the existing CityLearn framework, providing V2G and G2V simulation capabilities into the study of broader energy management strategies. Results validated EVLearn and its integration into CityLearn, where the impact of these strategies is highlighted through a comparative simulation scenario.

cs.MA↗

arXiv2024

EnergAIze: Multi Agent Deep Deterministic Policy Gradient for Vehicle to Grid Energy Management

This paper investigates the increasing roles of Renewable Energy Sources (RES) and Electric Vehicles (EVs). While indicating a new era of sustainable energy, these also introduce complex challenges, including the need to balance supply and demand and smooth peak consumptions amidst rising EV adoption rates. Addressing these challenges requires innovative solutions such as Demand Response (DR), energy flexibility management, Renewable Energy Communities (RECs), and more specifically for EVs, Vehicle-to-Grid (V2G). However, existing V2G approaches often fall short in real-world adaptability, global REC optimization with other flexible assets, scalability, and user engagement. To bridge this gap, this paper introduces EnergAIze, a Multi-Agent Reinforcement Learning (MARL) energy management framework, leveraging the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm. EnergAIze enables user-centric and multi-objective energy management by allowing each prosumer to select from a range of personal management objectives, thus encouraging engagement. Additionally, it architects' data protection and ownership through decentralized computing, where each prosumer can situate an energy management optimization node directly at their own dwelling. The local node not only manages local energy assets but also fosters REC wide optimization. The efficacy of EnergAIze was evaluated through case studies employing the CityLearn simulation framework. These simulations were instrumental in demonstrating EnergAIze's adeptness at implementing V2G technology within a REC and other energy assets. The results show reduction in peak loads, ramping, carbon emissions, and electricity costs at the REC level while optimizing for individual prosumers objectives.

cs.MA↗

arXiv2023

From Data to Action: Exploring AI and IoT-driven Solutions for Smarter Cities

The emergence of smart cities demands harnessing advanced technologies like the Internet of Things (IoT) and Artificial Intelligence (AI) and promises to unlock cities' potential to become more sustainable, efficient, and ultimately livable for their inhabitants. This work introduces an intelligent city management system that provides a data-driven approach to three use cases: (i) analyze traffic information to reduce the risk of traffic collisions and improve driver and pedestrian safety, (ii) identify when and where energy consumption can be reduced to improve cost savings, and (iii) detect maintenance issues like potholes in the city's roads and sidewalks, as well as the beginning of hazards like floods and fires. A case study in Aveiro City demonstrates the system's effectiveness in generating actionable insights that enhance security, energy efficiency, and sustainability, while highlighting the potential of AI and IoT-driven solutions for smart city development.

cs.LG↗

arXiv2023

Constrained Adversarial Learning for Automated Software Testing: a literature review

It is imperative to safeguard computer applications and information systems against the growing number of cyber-attacks. Automated software testing tools can be developed to quickly analyze many lines of code and detect vulnerabilities by generating function-specific testing data. This process draws similarities to the constrained adversarial examples generated by adversarial machine learning methods, so there could be significant benefits to the integration of these methods in testing tools to identify possible attack vectors. Therefore, this literature review is focused on the current state-of-the-art of constrained data generation approaches applied for adversarial learning and software testing, aiming to guide researchers and developers to enhance their software testing tools with adversarial testing methods and improve the resilience and robustness of their information systems. The found approaches were systematized, and the advantages and limitations of those specific for white-box, grey-box, and black-box testing were analyzed, identifying research gaps and opportunities to automate the testing tools with data generated by adversarial attacks.

cs.SE↗

arXiv2022

An IoT Cloud and Big Data Architecture for the Maintenance of Home Appliances

Billions of interconnected Internet of Things (IoT) sensors and devices collect tremendous amounts of data from real-world scenarios. Big data is generating increasing interest in a wide range of industries. Once data is analyzed through compute-intensive Machine Learning (ML) methods, it can derive critical business value for organizations. Powerfulplatforms are essential to handle and process such massive collections of information cost-effectively and conveniently. This work introduces a distributed and scalable platform architecture that can be deployed for efficient real-world big data collection and analytics. The proposed system was tested with a case study for Predictive Maintenance of Home Appliances, where current and vibration sensors with high acquisition frequency were connected to washing machines and refrigerators. The introduced platform was used to collect, store, and analyze the data. The experimental results demonstrated that the presented system could be advantageous for tackling real-world IoT scenarios in a cost-effective and local approach.

eess.SP↗

arXiv2022

A Low-Cost Multi-Agent System for Physical Security in Smart Buildings

Modern organizations face numerous physical security threats, from fire hazards to more intricate concerns regarding surveillance and unauthorized personnel. Conventional standalone fire and intrusion detection solutions must be installed and maintained independently, which leads to high capital and operational costs. Nonetheless, due to recent developments in smart sensors, computer vision techniques, and wireless communication technologies, these solutions can be integrated in a modular and low-cost manner. This work introduces Integrated Physical Security System (IP2S), a multi-agent system capable of coordinating diverse Internet of Things (IoT) sensors and actuators for an efficient mitigation of multiple physical security events. The proposed system was tested in a live case study that combined fire and intrusion detection in an industrial shop floor environment with four different sectors, two surveillance cameras, and a firefighting robot. The experimental results demonstrate that the integration of several events in a single automated system can be advantageous for the security of smart buildings, reducing false alarms and delays.

cs.CR↗

arXiv2022

A Multi-Policy Framework for Deep Learning-Based Fake News Detection

Connectivity plays an ever-increasing role in modern society, with people all around the world having easy access to rapidly disseminated information. However, a more interconnected society enables the spread of intentionally false information. To mitigate the negative impacts of fake news, it is essential to improve detection methodologies. This work introduces Multi-Policy Statement Checker (MPSC), a framework that automates fake news detection by using deep learning techniques to analyze a statement itself and its related news articles, predicting whether it is seemingly credible or suspicious. The proposed framework was evaluated using four merged datasets containing real and fake news. Long-Short Term Memory (LSTM), Gated Recurrent Unit (GRU) and Bidirectional Encoder Representations from Transformers (BERT) models were trained to utilize both lexical and syntactic features, and their performance was evaluated. The obtained results demonstrate that a multi-policy analysis reliably identifies suspicious statements, which can be advantageous for fake news detection.

cs.CL↗

arXiv2014

The higher spin generalization of the 6-vertex model with domain wall boundary conditions and Macdonald polynomials

The determinantal form of the partition function of the 6-vertex model with domain wall boundary conditions was given by Izergin. It is known that for a special value of the crossing parameter the partition function reduces to a Schur polynomial. Caradoc, Foda and Kitanine computed the partition function of the higher spin generalization of the 6-vertex model. In the present work it is shown that for a special value of the crossing parameter, referred to as the combinatorial point, the partition function reduces to a Macdonald polynomial.

math.CO↗

arXiv2013

Temperley-Lieb R-matrices from generalized Hadamard matrices

New sets of rank n-representations of Temperley-Lieb algebra TL_N(q) are constructed. They are characterized by two matrices obeying a generalization of the complex Hadamard property. Partial classifications for the two matrices are given, in particular when they reduce to Fourier or Butson matrices.

math-ph↗

arXiv2013

On some polynomials enumerating Fully Packed Loops configurations, evaluation at negative values

In this article, we are interested in the enumeration of Fully Packed Loops configurations on a grid with a given noncrossing matching. These quantities also appear as the groundstate components of the Completely Packed Loops model as conjectured by Razumov and Stroganov and recently proved by Cantini and Sportiello. When considering matchings with p nested arches these quantities are known to be polynomials. In a recent article, Fonseca and Nadeau conjectured some unexpected properties of these polynomials, suggesting that these quantities could be combinatorially interpreted even for negative p. Here, we prove some of these conjectures. Notably, we prove that for negative p we can factor the polynomials into two parts a "positive" one and a "negative" one. Also, a sum rules of the negative part is proven.

math.CO↗

arXiv2010

On some polynomials enumerating Fully Packed Loop configurations

We are interested in the enumeration of Fully Packed Loop configurations on a grid with a given noncrossing matching. By the recently proved Razumov--Stroganov conjecture, these quantities also appear as groundstate components in the Completely Packed Loop model. When considering matchings with p nested arches, these numbers are known to be polynomials in p. In this article, we present several conjectures about these polynomials: in particular, we describe all real roots, certain values of these polynomials, and conjecture that the coefficients are positive. The conjectures, which are of a combinatorial nature, are supported by strong numerical evidence and the proofs of several special cases. We also give a version of the conjectures when an extra parameter tau is added to the equations defining the groundstate of the Completely Packed Loop model.

math.CO↗