MAGANet: Achieving Combinatorial Generalization by Modeling a Group Action
24. července 2023
Řečníci
Geonho Hwang
Sprecher:in · 0 Follower:innen
Jaewoong Choi
Sprecher:in · 0 Follower:innen
Hyunsoo Cho
Sprecher:in · 0 Follower:innen
Myungjoo Kang
Sprecher:in · 0 Follower:innen
O prezentaci
Combinatorial generalization refers to the ability to collect and assemble various attributes from diverse data to generate novel unexperienced data. This ability is considered a necessary passing point for achieving human-level intelligence. To achieve this ability, previous unsupervised approaches mainly focused on learning the disentangled representation, such as the variational autoencoder. However, recent studies discovered that the disentangled representation is insufficient for combinatorial generalization and is not even correlated. In this regard, we propose a novel framework for data generation that can robustly generalize under these distribution shift situations. Instead of representing each data, our model discovers the fundamental transformation between a pair of data by simulating a group action. To test the combinatorial generalizability, we evaluated our model in two settings: Recombination-to-Element and Recombination-to-Range. The experiments demonstrated that our method has quantitatively and qualitatively superior generalizability and generates better images than traditional models.Combinatorial generalization refers to the ability to collect and assemble various attributes from diverse data to generate novel unexperienced data. This ability is considered a necessary passing point for achieving human-level intelligence. To achieve this ability, previous unsupervised approaches mainly focused on learning the disentangled representation, such as the variational autoencoder. However, recent studies discovered that the disentangled representation is insufficient for combinator…
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
Provably Convergent Schrödinger Bridge with Applications to Probabilistic Time Series Imputation
Später ansehen
Wei Deng, …
The Pulse of Ethical Machine Learning in Health
Später ansehen
Generalization on the Unseen, Logic Reasoning and Degree Curriculum
Später ansehen
Modeling Accurate Long Rollouts with Temporal Neural PDE Solvers
Später ansehen
