Contacts

Community

Rechercher

IT | EN | FR | عربي | GR

Corso Vittorio Emanuele II, 39 - Roma 0669207671

Contacts

Community

IT | EN | FR | عربي | GR

Ingénierie Informatique (Academic Year 2023/2024) - Sistemi Intelligenti

Sistemi artificiali adattivi

Leçon n. 1: Artificial adaptive systems The concept of AI What is artificial intelligence Simulation approach Emulation approach Information approach		Massimo Paolo Buscema
Leçon n. 2: Scientific knowledge workflow The procedure of scientific processes Data & Algorithms, Validation Anomalies - case study. Epistemological question Self organizing maps AWIT ANN Auto contractive map		Massimo Paolo Buscema
Leçon n. 3: Top down procedure vs bottom up procedure Rule-based systems and data driven systems The rational way The math way (top down) The AI way (bottom up) Artificial adaptive systems - evolutionary algorithms - population oriented Coding - classic tsp		Massimo Paolo Buscema
Leçon n. 4: Complicated systems & complex system Processes that require rules to function and Processes that create rules as they operate Features of complex systems		Massimo Paolo Buscema
Leçon n. 5: The information approach - first part Data base Types of data matrix Static matrix Dynamic matrix Iconic matrix Artificial adaptive systems (AAS)		Massimo Paolo Buscema
Leçon n. 6: The information approach - second part What analytics means Types of data analytics Representational analytics Graph analytics Simulation analytics Prescriptive analytics Space analytics Pixel analytics Diagnostic analytics Predictive analytics		Massimo Paolo Buscema
Leçon n. 7: Elements of theory of graphs Elements of theory of graphs and types of graphs From the adjacency matrix to the relationship matrix (D matrix) Weighted graph Types of graph structure & regular graph		Massimo Paolo Buscema
Leçon n. 8: Types of filter of weighted regular and complete graph The minimum spanning tree (MST) Hubness function (H function) and the delta h function Maximally regular graph (MRG) Graph IN-OUT		Massimo Paolo Buscema
Leçon n. 9: Graph indices and networks - first part Geometric & topological networks nodes indices The list of networks indices Examples of nodes indices		Massimo Paolo Buscema
Leçon n. 10: Graph indices and networks - second part graphs &networks Blatant & hidden networks example of a hidden structure Degree of separation Theory		Massimo Paolo Buscema
Leçon n. 11: Supervised validation protocol (random approach) - first part Assumptions Random distributions Reverse procedure 5x2 cross validation 3 steps protocol		Massimo Paolo Buscema
Leçon n. 12: Supervised validation protocol (random approach) - second part K fold cross validation Leave one out		Massimo Paolo Buscema
Leçon n. 13: Supervised validation protocol (inductive approach) - third part Inductive vs random split Inductive approach: test TWIST algorithm		Massimo Paolo Buscema
Leçon n. 14: Supervised metrics - first part Correct trend prediction Cross linear correlation Mean absolute error (MAE) Mean square error (MSE) Cross entropy loss function (CE)		Massimo Paolo Buscema
Leçon n. 15: Supervised metrics - second part R squared Accuracy-sensitivity-specificity-precision Confusion matrix for binary classes F1-score		Massimo Paolo Buscema
Leçon n. 16: Supervised metrics - third part Basic concepts of multinomial confusion matrix Multinomial confusion matrix applied		Massimo Paolo Buscema
Leçon n. 17: Unsupervised algorithms validation Test of reconstruction Test of MST similarity Test of MST accuracy applied to supervised data sets Test of correlation matrix accuracy applied to supervised data sets		Massimo Paolo Buscema
Leçon n. 18: The basic concepts of artificial adaptive systems The concept of AI Scientific knowledge workflow Top down procedures vs bottom up procedures Rules based systems and data driven systems Complicated systems & complex systems Processes that requires rules to function and processes that create rules as they operate The information approach What analytics means Elements of theory of graphs Supervised validation protocol Unsupervised algorithms validation		Massimo Paolo Buscema