DCC - UFMG
+55 31 3409 5860
ciia-saude@dcc.ufmg.br

Ana Bazzan


Ana Bazzan

Universidade Federal do Rio Grande do Sul
Informática Teórica

Associated researcher





Information extracted from Lattes platform


Last update: 2021/05/05


Degrees

Ph.D. Informatik na Karlsruher Institut für Technologie em 1997
M.Sc. Computação na Universidade Federal do Rio Grande do Sul em 1992
B.Sc. Engenharia na Universidade de São Paulo em 1984


Current projects

2019 a AtualMesoscopic Molecular Dynamics Simulations: Development of Models and Computational Strategies for Complex Structural Bioinformatics Problems

Integrantes: Marcio Dorn (coordenador), Ana Lucia Cetertich Bazzan.
2017 a Atual2UEI -- internet 2.0 e embarcada em veícUlos como fontes hEterogêneas de dados em cidades Inteligentes
Este projeto propõe métodos para recuperação e uso de dados de fontes heterogêneas, disponiveis atualmente na Internet, a fim prover informação tanto para autoridades e serviços de logística, como para o cidadão. No último caso, a finalidade é principalmente fornecer informações personalizadas (ou seja dentro do contexto do motorista), algo que hoje é pouco explorado. Tais fontes heterogêneas envolvem: texto (redes sociais como Twitter); boletins meteorológicos; eventos esportivos e culturais; imagens instantâneas de fluidez do trânsito; videos (webcams); comunicação interveicular e outras fontes ligadas com mobility Internet. Desta forma, como tipicamente é o caso na Internet 2.0, informações fluem nos dois sentidos e são aproveitadas por ambos, viajante e autoridade.
Integrantes: Ana Lucia Cetertich Bazzan (coordenador), Marcia Pasin, Jorge Aching.
2017 a AtualMultiagent Reinforcement Learning: Theoretical Investigations and Applications
O objetivo deste projeto é duplo: por um lado, serão conduzidas pesquisas em nível teórico visando expandir métodos para lidar com aprendizado por reforço multiagente. Por outro lado, estes métodos serão aplicados em problemas de larga escala do mundo real como escolha de rotas e controle semafórico, ambos no domínio de tráfego urbano.
Integrantes: Ana Lucia Cetertich Bazzan (coordenador), Bruno Castro da Silva.
2016 a AtualAprendizagem de Máquina Decentralizada: Uma Abordagem Baseada em Funções de Escolha Social
Uma suposição comumente adotada em tarefas de Aprendizado de Máquina (AM), notadamente as de classificação de padrões, é a de que o repositório de dados esteja todo ele centralizado e disponível quando da etapa de indução do modelo de estimação (classificador). Contudo, em muitas situações, tal suposição se mostra impraticável, haja vista a existência de restrições relativas à privacidade dos dados e aos altos custos de armazenamento, transmissão e processamento dos mesmos. Para lidar com esse cenário mais desafiador, novos sistemas de AM descentralizados vêm sendo desenvolvidos nos últimos anos, em que estimadores locais, encapsulados em agentes autônomos, passam a ser inferidos separadamente e combinados posteriormente em um modelo de consenso global. Nesse contexto específico, o presente projeto almeja investigar novas extensões de uma abordagem de classificação descentralizada recém-criada pelo nosso grupo, a qual se baseia na agregação das decisões dos agentes-classificadores via funções de escolha social ({\it social choice functions}). Mais precisamente, pretende-se estudar mais detalhadamente as propriedades de diferentes funções de escolha/bem-estar social existentes na literatura e analisar o impacto prático da sua adoção quando da combinação das saídas dos classificadores (heterogêneos) descentralizados. Ademais, pretende-se investigar novos esquemas hierárquicos de agregação, em que as funções de escolha/bem-estar social são usadas em camadas. Em vez de focar exclusivamente na tarefa de classificação binária, como feito originalmente, variantes mais complexas serão consideradas ao longo deste projeto, abarcando a classificação com múltiplas classes, a classificação multirrótulo e a classificação semi-supervisionada. Finalmente, será analisado o impacto da existência de ruído corrompendo os \rks\ gerados pelos classificadores sobre o desempenho da abordagem.
Integrantes: Ana Lucia Cetertich Bazzan (coordenador), Mariana Recamonde-Mendoza.
2016 a AtualINVESTIGAÇÃO SOBRE RESOLUÇÃO DE CONFLITOS E ELOS DE LIGAÇÃO EM REDES SOCIAIS DE EVENTOS HISTÓRICOS: UMA ABORDAGEM BASEADA EM REDES COMPLEXAS
Este projeto busca retomar o estudo de eventos da Idade Média através de uma perspectiva da teoria de redes (network theory), em particular redes complexas temporais. A ciência de redes complexas, que envolve as áreas de Matemática, Física e Ciência da Computação tem sido cada vez mais utilizada na análise de textos históricos e literários, uma vez que permite uma visão ampla e fundamentada da interconectividade de personagens históricos durante longos períodos de tempo.
Integrantes: Ana Lucia Cetertich Bazzan (coordenador), DAHMEN, S, PRADO, SANDRA D., CYBELE CROSSETTI DE ALMEIDA.

Current applied research projects

See all projects in Lattes

Recent publications

Articles in journals

Accelerating route choice learning with experience sharing in a commuting scenario: An agent-based approach
2021. AI COMMUNICATIONS.
Contribuições de aprendizado por reforço em escolha de rota e controle semafórico
2021. ESTUDOS AVANÇADOS (ONLINE).
Sharing diverse information gets driver agents to learn faster: an application in en route trip building
2021. PEERJ COMPUTER SCIENCE.
Reinforcement learning vs. rule-based adaptive traffic signal control: A Fourier basis linear function approximation for traffic signal control
2021. AI COMMUNICATIONS.
Hierarchical Traffic Signal Optimization Using Reinforcement Learning and Traffic Prediction with Long-Short Term Memory
2021. EXPERT SYSTEMS WITH APPLICATIONS.
Toll-based reinforcement learning for efficient equilibria in route choice
2020. KNOWLEDGE ENGINEERING REVIEW.
Quantitatively assessing the benefits of model-driven development in agent-based modeling and simulation
2020. SIMULATION MODELLING PRACTICE AND THEORY.
Similar Yet Different: the Structure of Social Networks of Characters in Seinfeld, Friends, How I Met Your Mother, and The Big Bang Theory
2020. REVISTA DE INFORMÁTICA TEÓRICA E APLICADA: RITA.
A comparative evaluation of aggregation methods for machine learning over vertically partitioned data
2020. EXPERT SYSTEMS WITH APPLICATIONS.
Experience classification for transfer learning in traffic signal control
2020. THE JOURNAL OF SUPERCOMPUTING (DORDRECHT. ONLINE).
A New Microscopic Approach to Traffic Flow Classification Using a Convolutional Neural Network Object Detector and a Multi-Tracker Algorithm
2020. IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS.
Gendered Networks and Communicability in Medieval Historical Narratives Advances in Complex Systems
2020. ADVANCES IN COMPLEX SYSTEMS.
How hard is for agents to learn the user equilibrium? Characterizing traffic networks by means of entropy
2020. ADVANCES IN COMPLEX SYSTEMS.
Identifying traffic conditions from non-traffic related sources
2020. Journal of Intelligent Transportation Systems.
I will be there for you: clique, character centrality, and community detection in Friends
2020. COMPUTATIONAL AND APPLIED MATHEMATICS.
Aligning individual and collective welfare in complex socio-technical systems by combining metaheuristics and reinforcement learning
2019. ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE.
Combining adaptation at supply and demand levels in microscopic traffic simulation: a multiagent learning approach
2019. Transportation Research Procedia.
Incentive-based search for equilibria in boolean games
2019. Constraints (Dordrecht. Online).
A Reinforcement Learning-based Multi-Agent Framework applied for solving routing and scheduling problems
2019. EXPERT SYSTEMS WITH APPLICATIONS.
A context-aware convention formation framework for large-scale networks
2018. AUTONOMOUS AGENTS AND MULTI-AGENT SYSTEMS.
Optimal Electric Vehicle Fast Charging Station Placement Based on Game Theoretical Framework
2018. IEEE Transactions on Intelligent Transportation Systems.
Model-driven agent-based simulation development: A modeling language and empirical evaluation in the adaptive traffic signal control domain
2018. SIMULATION MODELLING PRACTICE AND THEORY.
Solving task allocation problem in multi Unmanned Aerial Vehicles systems using Swarm intelligence
2018. ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE.
Analysing the impact of travel information for minimising the regret of route choice
2018. TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES.
Liar liar, pants on fire; or how to use subjective logic and argumentation to evaluate information from untrustworthy sources
2017. ARTIFICIAL INTELLIGENCE REVIEW (DORDRECHT. ONLINE).
Social choice functions: A tool for ranking variables involved in action plans against road noise
2016. Journal of Environmental Management.
TEMPORAL NETWORK ANALYSIS OF LITERARY TEXTS
2016. Advances in Complex Systems.
Social choice in distributed classification tasks: Dealing with vertically partitioned data
2016. Information Sciences.
Opportunities for multiagent systems and multiagent reinforcement learning in traffic control
2009. Autonomous Agents and Multi-Agent Systems.
A Distributed Approach for Coordination of Traffic Signal Agents
2005. Autonomous Agents and Multi-Agent Systems.
Case studies on the Braess Paradox: Simulating route recommendation and learning in abstract and microscopic models
2005. Transportation Research. Part C, Emerging Technologies.
The impact of real-time information in a two-route scenario using agent-based simulation
2002. Transportation Research. Part C, Emerging Technologies.
Decision dynamics in a traffic scenario
2000. Physica. A (Print).

Papers in conferences

Minimum-Delay Adaptation in Non-Stationary Reinforcement Learning via Online High-Confidence Change-Point Detection
2021. 20th International Conference on Autonomous Agents and MultiAgent System.
Detecting Communities in Networks: a Decentralized Approach Based on Multiagent Reinforcement Learning
2020. 2020 IEEE Symposium Series on Computational Intelligence (SSCI).
Inducing selfish agents towards social efficient solutions
2020. Symposium on Knowledge Discovery, Mining and Learning.
Accelerating learning of route choices with C2I: a preliminary investigation
2020. Symposium on Knowledge Discovery, Mining and Learning.
Experience Sharing in a Traffic Scenario
2020. Workshop Agents in Traffic and Transportation.
A reinforcement learning approach with Fourier basis linear function approximation for traffic signal control
2020. Workshop Agents in Traffic and Transportation.
Preventive maintenance tool for offshore units using decentralized heterogeneous data
2020. Rio Oil & Gas Expo and Conference.
Dynamic Correlation-based Feature Selection for Feature Drifts in Data Streams
2019. BRACIS.
Identifying traffic event types from Twitter by Multi-label Classification
2019. BRACIS.
O QUE É RELEVANTE NA PREDIÇÃO DAS CONDIÇÕES DO TRÁFEGO A PARTIR DE DADOS DE TWITTER E OUTRAS FONTES HETEROGÊNEAS? UM ESTUDO PRELIMINAR SOBRE A CIDADE DE PORTO ALEGRE
2019. Congresso de Pesquisa e Ensino em Transportes.
Using Information from Heterogeneous Sources and Machine Learning in Intelligent Transportation Systems
2019. 2019 IEEE 15th International Conference on Intelligent Computer Communication and Processing (ICCP).
Towards Designing Optimal Reward Functions in Multi-Agent Reinforcement Learning Problems
2018. International Joint Conference on Neural Networks (IJCNN).
Accelerating the Computation of Solutions in Resource Allocation Problems Using an Evolutionary Approach and Multiagent Reinforcement Learning
2018. International Conference on the Applications of Evolutionary Computation.
Adaptive Incremental Gaussian Mixture Network for Non-Stationary Data Stream Classification
2018. 2018 International Joint Conference on Neural Networks (IJCNN).
Comparing Multi-Armed Bandit Algorithms and Q-learning for Multiagent Action Selection: a Case Study in Route Choice
2018. 2018 International Joint Conference on Neural Networks (IJCNN).
Co-Adaptive Reinforcement Learning in Microscopic Traffic Systems
2018. 2018 IEEE Congress on Evolutionary Computation (CEC).
Supporting the Development of Agent-Based Simulations: A DSL for Environment Modeling
2017. 2017 IEEE 41st Annual Computer Software and Applications Conference (COMPSAC).
Learning to Minimise Regret in Route Choice
2017. International Joint Conference on Autonomous Agents and Multiagents Systems (AAMAS).
Understanding the Behaviour of Learning-Based BDI Agents in the Braess' Paradox
2017. German Conference on Multiagent System Technologies.
Social-Training: Ensemble Learning with Voting Aggregation for Semi-supervised Classification Tasks
2017. 2017 Brazilian Conference on Intelligent Systems (BRACIS).
Comparing Two Multiagent Reinforcement Learning Approaches for the Traffic Assignment Problem
2017. 2017 Brazilian Conference on Intelligent Systems (BRACIS).
On Estimating Regret Locally and Learning From It in Route Choice
2016. Workshop on Agents in Traffic and Transportation.
Using Topological Statistics to Bias and Accelerate Route Choice: preliminary findings in artificial and real-world road networks
2016. Workshop on Agents in Traffic and Transportation.
Efficient local search in traffic assignment
2016. 2016 IEEE Congress on Evolutionary Computation (CEC).
Combining Car-to-Infrastructure Communication and Multi-Agent Reinforcement Learning in Route Choice
2016. Workshop on Agents in Traffic and Transportation.
Autonomous car and ride sharing: flexible road trains
2016. the 24th ACM SIGSPATIAL International Conference.
A multiagent reinforcement learning approach to en-route trip building
2016. 2016 International Joint Conference on Neural Networks (IJCNN).
A multiagent solution to overcome selfish routing in transportation networks
2016. 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC).
A bi-objective method of traffic assignment for electric vehicles
2016. 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC).

Extended abstracts in conferences

Model-Driven Engineering in Agent-based Modeling and Simulation: a Case Study in the Traffic Signal Control Domain
2017. International Joint Conference on Autonomous Agents and Multiagents Systems (AAMAS).
A Flexible Approach for Designing Optimal Reward Functions
2017. International Joint Conference on Autonomous Agents and Multiagents Systems (AAMAS).
Synergies between evolutionary computation and multiagent reinforcement learning
2017. the Genetic and Evolutionary Computation Conference Companion.
Hybrid Evolutionary and Reinforcement Learning Approach to Accelerate Traffic Assignment (extended abstract)
2015. International Joint Conference on Autonomous Agents and Multiagent Systems (AAMAS).
Getting You Faster to Work - A Genetic Algorithm Approach to the Traffic Assignment Problem
2014. Genetic and Evolutionary Computation Conference (GECCO).

Abstracts in conferences

Recommending Travel Packages Upon Distributed Knowledge
2007. AAAI.
ITSUMO: an Intelligent Transportation System for Urban Mobility
2006. 5th International Joint Conference On Autonomous Agents And Multiagent Systems (AAMAS 2006).
A Case Study On The Role Of Information For Implicit Coordination
2004. The Third International Joint Conference on Autonomous Agents and Multiagent Systems.
Route Decision Behaviour in a Commuting Scenario: Simple Heuristics Adaptation and Effect of Traffic Forecast
2003. Second UCLA Lake Arrowhead Conference on Human Complex Systems.
Simulation of Adaptive Agents: Learning Heuristics for Route Choice in a Commuter Scenario
2002. First Int. Joint Conf. on Autonomous Agents and Multiagent Systems (AAMAS 2002).

See all publications in Lattes

Current students

MS

João Vitor Bruniera Labres. em definição. Início: 2020. Universidade Federal do Rio Grande do Sul (Orientador principal)
Taylor de Oliveira Antes. em definição. Início: 2019. Universidade Federal do Rio Grande do Sul (Orientador principal)
Candy Alexandra Huanca-Anquise. em definição. Início: 2019. Universidade Federal do Rio Grande do Sul (Orientador principal)
Rhuam Sena Estevam. em definicao. Início: 2018. Universidade Federal do Rio Grande do Sul (Orientador principal)

PhD

Lucas Nunes Alegre. Model-Based Multi-Objective Deep Reinforcement Learning. Início: 2021. Universidade Federal do Rio Grande do Sul (Orientador principal)
Jorge C. Chamby-Diaz. em definicao. Início: 2018. Universidade Federal do Rio Grande do Sul (Orientador principal)

See all students in Lattes