Self-supervised information retrieval: a novel approach based on Deep Metric Learning and Neural Language Models

Most of the existing open-source search engines, utilize keyword or tf-idf based techniques to find relevant documents and web pages relative to an input query. Although these methods, with the help of a page rank or knowledge graphs, proved to be effective in some cases, they often fail to retrieve relevant instances for more complicated queries that would require a semantic understanding to be exploited. In this Thesis, a self-supervised information retrieval system based on transformers is employed to build a semantic search engine over the library of Gruppo Maggioli company. Semantic search or search with meaning can refer to an understanding of the query, instead of simply finding words matches and, in general, it represents knowledge in a way suitable for retrieval. We chose to investigate a new self-supervised strategy to handle the training of unlabeled data based on the creation of pairs of ’artificial’ queries and the respective positive passages. We claim that by removing the reliance on labeled data, we may use the large volume of unlabeled material on the web without being limited to languages or domains where labeled data is abundant.

Knowledge-based Chord Embeddings

This thesis develops AI methods as a contribution to computational musicology, an interdisciplinary field that studies music with computers. In systematic musicology a composition is defined as the combination of harmony, melody and rhythm. According to de La Borde, harmony alone "merits the name of composition". This thesis focuses on analysing the harmony from a computational perspective. We concentrate on symbolic music representation and address the problem of formally representing chord progressions in western music compositions. Informally, chords are sets of pitches played simultaneously, and chord progressions constitute the harmony of a composition. Our approach combines ML techniques with knowledge-based techniques. We design and implement the Modal Harmony ontology (MHO), using OWL. It formalises one of the most important theories in western music: the Modal Harmony Theory. We propose and experiment with different types of embedding methods to encode chords, inspired by NLP and adapted to the music domain, using both statistical (extensional) knowledge by relying on a huge dataset of chord annotations (ChoCo), intensional knowledge by relying on MHO and a combination of the two. The methods are evaluated on two musicologically relevant tasks: chord classification and music structure segmentation. The former is verified by comparing the results of the Odd One Out algorithm to the classification obtained with MHO. Good performances (accuracy: 0.86) are achieved. We feed a RNN for the latter, using our embeddings. Results show that the best performance (F1: 0.6) is achieved with embeddings that combine both approaches. Our method outpeforms the state of the art (F1 = 0.42) for symbolic music structure segmentation. It is worth noticing that embeddings based only on MHO almost equal the best performance (F1 = 0.58). We remark that those embeddings only require the ontology as an input as opposed to other approaches that rely on large datasets.

Distribuzione spazio temporale dei pesci demersali sulla piattaforma continentale del Golfo di Antalya

Le variabili ambientali e lo sfruttamento della pesca sono dei possibili fattori nel determinare la struttura della comunità demersale. L’area di studio è il Golfo di Antalya, con un area aperta ed una chiusa ad ogni attività di pesca, il periodo di studio ha coperto tre stagioni (primavera, estate, autunno). Lo scopo è quello di delineare un quadro generale sulla distribuzione spaziale e temporale delle risorse alieutiche demersali in quest’area. In questo lavoro di tesi la PCA è stata usata al fine di determinare le variabili ambientali (ossigeno, salinità, temperatura, pH, materia sospesa) che determinano maggiormente la differenza tra le stazioni, tecniche di analisi multivariata hanno invece indagato una possibile variazione su scala spaziale e temporale dei parametri abiotici. La Cluster Analysis effettuata sui dati di abbondanza ha delineato quattro raggruppamenti principali, due ad una profondità minore di 100 m e due ad una profondità maggiore (40% di similarità). Questi risultati sono confermati dall’analisi MDS. L’analisi SIMPER ha messo in evidenza le specie che maggiormente incidono sulla differenza tra strati di profondità. Gli indici di biodiversità sono stati calcolati per indagare la diversità e la variabilità temporale e spaziale della comunità demersale. Due procedure la BIO-ENV e la DistLM (Distance-based linear models) sono state effettuate per individuare le variabili abiotiche che potrebbero essere responsabili dei diversi raggruppamenti nella struttura del popolamento demersale. Le specie commerciali: Mullus barbatus, Upeneus moluccensis, Upeneus pori sono state prese come oggetto per la ricerca di possibili effetti della pesca a livello di popolazione. Per i dati di abbondanza e di biomassa di queste specie è stata eseguita l’analisi multivariata MANOVA (Multivariate Analysis of Variance) al fine di trovare eventuali variazioni dovute ai fattori profondità, stagione e transetto. Per ogni specie è stata valutata la sex ratio. Il metodo Bhattacharya ha permesso di determinare le classi di età e la loro abbondanza. In ultimo la relazione peso-lunghezza è stata ricavata separatamente per gli individui maschi e femmine al fine di determinare il tipo di crescita per ogni sesso.

Synthesis, Characterization, and Applications of a Novel Cu(II)-MOF Based on a Propargylcarbamate-Functionalized Isophthalate Ligand

This work describes the synthesis of a propargylcarbamate-functionalized isophthalate ligand and its use in the solvothermal preparation of a new copper(II)-based metal organic framework named [Cu(1,3-YBDC)]ˑxH2O (also abbreviated as Cu-MOF. The characterization of this compound was performed using several complementary techniques such as infrared (ATR-FTIR) and Raman spectroscopy, X-ray powder diffraction spectroscopy (PXRD), scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS), atomic absorption spectroscopy (AAS) as well as thermal and surface area measurements. Synchrotron X-ray diffraction analysis revealed that this MOF contains a complex network of 5-substituted isophthalate anions bound to Cu(II) centers, arranged in pairs within paddlewheel (or “Chinese lantern”) structure with a short Cu…Cu distance of 2.633 Å. Quite unexpectedly, the apical atom in the paddlewheel structure belongs to the carbamate carbonyl oxygen atom. Such extra coordination by the propargylcarbamate groups drastically reduces the MOF porosity, a feature that was also confirmed by BET measurements. Indeed, its surface area was determined to be low (14.5 ± 0.8 m2/g) as its total pore volume (46 mm3/g). Successively the Cu-MOF was treated with HAuCl4 with the aim of studying the ability of the propargylcarbamate functionality to capture the Au(III) ion and reduce it to Au(0) to give gold nanoparticles (AuNPs). The overall amount of gold retained by the Cu-MOF/Au was determined by AAS while the amount of gold and its oxidation state on the surface of the MOF was studied by XPS. A glassy carbon (GC) electrode was drop-casted with a Cu-MOF suspension to electrochemically characterize the material through cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). The performance of the modified electrodes towards nitrite oxidation was tested by CV and chronoamperometry.

Graph-based User Scheduling for MIMO LEO-based Satellite Communication Systems

The study of the user scheduling problem in a Low Earth Orbit (LEO) Multi-User MIMO system is the objective of this thesis. With the application of cutting-edge digital beamforming algorithms, a LEO satellite with an antenna array and a large number of antenna elements can provide service to many user terminals (UTs) in full frequency reuse (FFR) schemes. Since the number of UTs on-ground are many more than the transmit antennas on the satellite, user scheduling is necessary. Scheduling can be accomplished by grouping users into different clusters: users within the same cluster are multiplexed and served together via Space Division Multiple Access (SDMA), i.e., digital beamforming or Multi-User MIMO techniques; the different clusters of users are then served on different time slots via Time Division Multiple Access (TDMA). The design of an optimal user grouping strategy is known to be an NP-complete problem which can be solved only through exhaustive search. In this thesis, we provide a graph-based user scheduling and feed space beamforming architecture for the downlink with the aim of reducing user inter-beam interference. The main idea is based on clustering users whose pairwise great-circle distance is as large as possible. First, we create a graph where the users represent the vertices, whereas an edge in the graph between 2 users exists if their great-circle distance is above a certain threshold. In the second step, we develop a low complex greedy user clustering technique and we iteratively search for the maximum clique in the graph, i.e., the largest fully connected subgraph in the graph. Finally, by using the 3 aforementioned power normalization techniques, a Minimum Mean Square Error (MMSE) beamforming matrix is deployed on a cluster basis. The suggested scheduling system is compared with a position-based scheduler, which generates a beam lattice on the ground and randomly selects one user per beam to form a cluster.