Laboratory seminar December 11

LATNA  Laboratory scientific seminar will take place on Wednesday, December 11, at 16:00, in the auditorium 401 in the HSE building at Rodionova Street (usual day, place and time of the seminar).
Speaker: Liubov Tupikina (MIPT and ITMO).
Title: Dissecting embedding methods: learning higher-order structures from data
Abstract: Active area of research in computer science is the theory of manifold learning and finding lower-dimensional manifold representation on how we can learn geometry from data for providing better quality curated datasets. However geometric machine learning and deep learning methods for data learning often include set of assumptions on the geometry of the feature space. Some of these assumptions include pre-selected metrics on the feature space, usage of the underlying graph structure, which encodes the data points proximity. However, the later assumption of using a graph as the underlying discrete structure, encodes only the binary pairwise relations between data points, restricting ourselves from capturing more complex higher-order relationships, which are often often present in various systems. These assumptions on the data together with data being discrete and finite may cause some generalisation, which may create wrong interpretations of the data and models, which produce the embeddings of data itself, such as BERT and others.
In the talk we will talk about the alternative framework characterize the embedding methods using the higher-order structures. For this we explore the underlying graph assumption substituting it with the hypergraph structure. We also demonstrate the embedding characterization on the usecase of the example of higher order data.