Group-invariant max filtering
This interesting paper speaks to a common problem in modern machine learning when the number of data points is small.\N\NIf the number of data points is large then one is able to use a variety of modern tools to deal with matters such as the curse of dimensionality and estimates of dimension for example.\N\NThese techniques range for example from manifold learning to nonlinear dimension reduction and many others. Recent papers have dealt with data points of small cardinality by various techniques for example to controlling local neighborhoods of the points.\N\NThis paper proposes an interesting idea to deal with point sets of small cardinality. One idea to deal with this problem is to introduce small distortions in the data whose actions correspond to that of a group meaning the ground truth exhibits known symmetries. This encourages the data set to learn such actions, often symmetries, and they of course increase the cardinality of the set of data points. Although this idea has proved successful in many applications, it is often inefficient to train a large, symmetry-agnostic model to find a highly symmetric function. This then naturally begs the question of why not use a model which accounts for known symmetries and this is the subject of invariant machine learning.\N\NThe authors do the following. Let us be given a group \(G\) of linear isometries and a real inner product space \(V\). The authors then construct what they call max filters which are a family of \(G\)-invariant real-valued functions on \(V\). For a special case to give some insight. Take \(V = R^d\) and \(G\) finite. Then a a suitable max filter bank separates orbits, and is bilipschitz in the quotient metric. In the case where \(V = L_2(R^d )\) and \(G\) is the group of translation operators, a max filter shows stability to diffeomorphic distortion. This is similar to the scattering transform which has been introduced by Mallat. The authors show that max filters are well suited for various classification tasks, both in practice and theory.\N\NThe paper is well written with an excellent set of references.
- Almost everywhere injectivity conditions for the matrix recovery problem
- An algebraic characterization of injectivity in phase retrieval
- Characterization of the subdifferential of some matrix norms
- Color-coding
- Complete set of translation invariant measurements with Lipschitz bounds
- Computational invariant theory. With two appendices by Vladimir L. Popov and an addendum by Nobert A. Campo and Vladimir L. Popov
- Convex Analysis
- Degree bound for separating invariants of abelian groups
- Dihedral Multi-Reference Alignment
- Estimation under group actions: recovering orbits from invariants
- Extensions of Lipschitz mappings into a Hilbert space
- Fourier integrals in classical analysis
- Generalized phase retrieval: measurement number, matrix recovery and beyond
- Geometric Invariant Theory
- Group invariant scattering
- scientific article; zbMATH DE number 47220 (Why is no real title available?)
- scientific article; zbMATH DE number 7306924 (Why is no real title available?)
- Image manifolds which are isometric to Euclidean space
- Improved recovery guarantees for phase retrieval from coded diffraction patterns
- Invertibility and robustness of phaseless reconstruction
- Lipschitz analysis of generalized phase retrievable matrix frames
- Lower Lipschitz bounds for phase retrieval from locally supported measurements
- On Lipschitz Bounds of General Convolutional Neural Networks
- On signal reconstruction without phase
- On sparse reconstruction from Fourier and Gaussian measurements
- Phase recovery, MaxCut and complex semidefinite programming
- Phase retrieval from power spectra of masked signals
- Phase retrieval in infinite-dimensional Hilbert spaces
- Phase retrieval via matrix completion
- Phase retrieval via Wirtinger flow: theory and algorithms
- Phase retrieval with polarization
- Phaselift: exact and stable signal recovery from magnitude measurements via convex programming
- Saving phase: injectivity and stability for phase retrieval
- Separating monomials for diagonalizable actions
- Solving Random Quadratic Systems of Equations Is Nearly as Easy as Solving Linear Systems
- Stable optimizationless recovery from phaseless linear measurements
- Stable phase retrieval with low-redundancy frames
- Stable separation of orbits for finite abelian group actions
- Suprema of chaos processes and the restricted isometry property
- The restricted isometry property for time-frequency structured random matrices
- The sample complexity of multireference alignment
- Über die Theorie der algebraischen Formen.
- Functions on symmetric matrices and point clouds via lightweight invariant features from Galois theory
- A max filtering local stability theorem with application to weighted phase retrieval and cryo-EM
- Letter to the editor. Instance optimality in phase retrieval
- Coarse embeddings of quotients by finite group actions
- Separating polynomial invariants over non-closed fields of finite abelian groups
- A transversality theorem for semi-algebraic sets with application to signal recovery from the second moment and cryo-EM
This page was built for publication: Group-invariant max filtering
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6975880)
