M.Sc. in Artificial Intelligence

Curricula & syllabus

Sem I

Knowledge representation and reasoning

Knowledge representation in a model. First order and higher order models. Non-monotonic models. Temporal models. Logical models of higher order. Frame systems. Representing structured knowledge. Description logics and ontologies. Constraint-based representation and associated languages. Systems for maintaining data consistency. Bayesian networks. Propagation of trust. Plan representation and advanced techniques for automatic planning. Multi-paradigm representations and reasoning. Real life applications and usage of knowledge representation and automatic reasoning techniques.

Type systems and functional programming

Untyped Lambda Calculus. Recursion and fixed-point combinators.Operational semantics as a language specification tool. Typed Lambda Calculus. The "type system" concept. Salient properties: progress and preservation, type erasure and typability. Strong normalization in the Simple Typed Lambda Calculus (STLC). Extending STLC with particular monomorphic types: Boolean, natural numbers, product. Recursive types: lists. Parameterized polymorphism. System F. Type reconstruction: unification, principal types, typing rules. Higher-order types: "kinds". The Fw system. Particular problems: types, classes and functors in Haskell. Theoretical approaches of types: algebraic, constructivist logic, categorical.

Data Mining

Introduction in Data mining. Data preprocessing. Association Rules & Sequential Patterns. Supervised learning. Unsupervised learning – Clustering. Partially supervised learning. Information integration. Link analysis. Data warehousing. Dimensional modeling. Building a data warehouse.

1 Elective

Research activities

Sem II

Multi-agent systems

Agents and multi-agent systems. Architectures for cognitive and reactive agents. Communication languages and protocols for MAS. Coordination for solving tasks. Distributed planning in MAS. Negotiation techniques and protocols. Learning in MAS. Agent oriented programming. MAS platforms. Applications of multi-agent systems. Personal and Internet agents.

Natural language processing

Introduction in Natural Language Processing. Phonetics and phonology. Finite state transducers, two level morphology, paradigmatic morphology, Stemming and lemmatization. Corpus linguistics. Hidden Markov Models; Naïve Bayes method with applications in NLP. Different classes of grammatical formalisms for natural language. Unification grammars, chart parsing, Earley și CKY algorithms. Part of Speech Tagging Case grammars, Ontologies, Sense disambiguation. LSA, pLSA, LDA. Pragmatics and discourse analysis. Coreferences. Rhetorical schemas and natural language generation. Polyphonic theory. Conversation analysis

Symbolic and statistical learning

Introductive elements of machine learning, statistics, information theory and decision theory. Linear models for regressions. Linear models for classifications. Kernel methods and Gaussian processes. Sparse kernel methods (Support vector machines and Relevance vector machines). Bayesian methods and graphical methods. Expectation maximization. Principal components analysis and Independent component analysis. Hidden Markov models

1 Elective

Research activities

Sem III

Self-organizing systems

An introduction to self-organizing systems. Bio-inspired self-organizing systems. Self-organizing systems used in economy. Ant Colony Optimization. The social organism.

Elements of Evolutionary Computation. Elements of social psychology. Culture in theory and practice. Thinking as a social process. Particle Swarm. Particle Swarm Optimization.

Neural networks

Connectionist paradigm. Rules and learning algorithms of neural networks with feed forward. Universal function approximators feed forward propagation multilayer networks. Recursive Hopfield Networks. Boltzmann Machines. Self-organization principle of and supervised learning. Broomhead & Lowe networks with radial or elliptical basis functions. Cascor Neural Networks. Extracting knowledge from neural networks. Evolutionary intelligent agents to implement neural networks.Social learning.

Software Verification and Validation

Introduction: Software development. Software development methods and models. Extreme programming. Requirements analysis; user specifications; UML. Software testing. Testing techniques and methodologies. Software verification. Methods of software specification. The predictability of software development.

1 Elective

Research activities

Sem IV

Research activities and development of Master Thesis

Some Master of Science and research thesis of previous years:

Argumentation in Ambient Intelligence

Properties of Extension-Based Semantics

Normative Multi-Agent Systems with BDI Agents

Formalizing contexts for modeling relationships in MAS

Using a Social Trust Model to Secure

Routing in a Wireless Sensor Network

Microgrid Controlled by Multi-agent Systems