FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2888891189> ?p ?o ?g. }

• W2888891189 abstract "One of the most fundamental problems in machine learning is to compare examples: Given a pair of objects we want to return a value which indicates degree of (dis)similarity. Similarity is often task specific, and pre-defined distances can perform poorly, leading to work in metric learning. However, being able to learn a similarity-sensitive distance function also presupposes access to a rich, discriminative representation for the objects at hand. In this dissertation we present contributions towards both ends. In the first part of the thesis, assuming good representations for the data, we present a formulation for metric learning that makes a more direct attempt to optimize for the k-NN accuracy as compared to prior work. We also present extensions of this formulation to metric learning for kNN regression, asymmetric similarity learning and discriminative learning of Hamming distance. In the second part, we consider a situation where we are on a limited computational budget i.e. optimizing over a space of possible metrics would be infeasible, but access to a label aware distance metric is still desirable. We present a simple, and computationally inexpensive approach for estimating a well motivated metric that relies only on gradient estimates, discussing theoretical and experimental results. In the final part, we address representational issues, considering group equivariant convolutional neural networks (GCNNs). Equivariance to symmetry transformations is explicitly encoded in GCNNs; a classical CNN being the simplest example. In particular, we present a SO(3)-equivariant neural network architecture for spherical data, that operates entirely in Fourier space, while also providing a formalism for the design of fully Fourier neural networks that are equivariant to the action of any continuous compact group." @default.
• W2888891189 created "2018-09-07" @default.
• W2888891189 creator A5047292252 @default.
• W2888891189 date "2018-08-30" @default.
• W2888891189 modified "2023-09-27" @default.
• W2888891189 title "Discriminative Learning of Similarity and Group Equivariant Representations." @default.
• W2888891189 cites W112688168 @default.
• W2888891189 cites W1513167324 @default.
• W2888891189 cites W1540596182 @default.
• W2888891189 cites W1545302199 @default.
• W2888891189 cites W1593916594 @default.
• W2888891189 cites W1635256758 @default.
• W2888891189 cites W190437827 @default.
• W2888891189 cites W1912570122 @default.
• W2888891189 cites W1964357740 @default.
• W2888891189 cites W1966156785 @default.
• W2888891189 cites W1975721872 @default.
• W2888891189 cites W1978930761 @default.
• W2888891189 cites W1994906459 @default.
• W2888891189 cites W2002276939 @default.
• W2888891189 cites W2017977879 @default.
• W2888891189 cites W202794677 @default.
• W2888891189 cites W2029538739 @default.
• W2888891189 cites W2059507684 @default.
• W2888891189 cites W2067952552 @default.
• W2888891189 cites W2070771761 @default.
• W2888891189 cites W2071111253 @default.
• W2888891189 cites W2076063813 @default.
• W2888891189 cites W2084363474 @default.
• W2888891189 cites W2087347434 @default.
• W2888891189 cites W2100495367 @default.
• W2888891189 cites W2104489082 @default.
• W2888891189 cites W2104752854 @default.
• W2888891189 cites W2106053110 @default.
• W2888891189 cites W2107790757 @default.
• W2888891189 cites W2112796928 @default.
• W2888891189 cites W2113945798 @default.
• W2888891189 cites W2117154949 @default.
• W2888891189 cites W2121713321 @default.
• W2888891189 cites W2122018816 @default.
• W2888891189 cites W2124984007 @default.
• W2888891189 cites W2126221001 @default.
• W2888891189 cites W2127218421 @default.
• W2888891189 cites W2133296809 @default.
• W2888891189 cites W2134207659 @default.
• W2888891189 cites W2141864905 @default.
• W2888891189 cites W2147196093 @default.
• W2888891189 cites W2147800946 @default.
• W2888891189 cites W2150120413 @default.
• W2888891189 cites W2151103935 @default.
• W2888891189 cites W2153693853 @default.
• W2888891189 cites W2158139921 @default.
• W2888891189 cites W2160615084 @default.
• W2888891189 cites W2161627023 @default.
• W2888891189 cites W2163490846 @default.
• W2888891189 cites W2167383966 @default.
• W2888891189 cites W2171050905 @default.
• W2888891189 cites W2173027866 @default.
• W2888891189 cites W2185466002 @default.
• W2888891189 cites W2190691619 @default.
• W2888891189 cites W2398080148 @default.
• W2888891189 cites W2427881153 @default.
• W2888891189 cites W252252322 @default.
• W2888891189 cites W2530741948 @default.
• W2888891189 cites W2558460151 @default.
• W2888891189 cites W2563709626 @default.
• W2888891189 cites W2576915720 @default.
• W2888891189 cites W2582199702 @default.
• W2888891189 cites W2621199038 @default.
• W2888891189 cites W2751473119 @default.
• W2888891189 cites W2751706698 @default.
• W2888891189 cites W2783378529 @default.
• W2888891189 cites W2788775653 @default.
• W2888891189 cites W2810026216 @default.
• W2888891189 cites W2810443692 @default.
• W2888891189 cites W2949117887 @default.
• W2888891189 cites W2951128674 @default.
• W2888891189 cites W2951770173 @default.
• W2888891189 cites W2952054889 @default.
• W2888891189 cites W2952433032 @default.
• W2888891189 cites W2963021451 @default.
• W2888891189 cites W2963504932 @default.
• W2888891189 cites W2963829960 @default.
• W2888891189 cites W2964121744 @default.
• W2888891189 cites W2964321699 @default.
• W2888891189 cites W3214736482 @default.
• W2888891189 cites W637153065 @default.
• W2888891189 hasPublicationYear "2018" @default.
• W2888891189 type Work @default.
• W2888891189 sameAs 2888891189 @default.
• W2888891189 citedByCount "1" @default.
• W2888891189 countsByYear W28888911892020 @default.
• W2888891189 crossrefType "posted-content" @default.
• W2888891189 hasAuthorship W2888891189A5047292252 @default.
• W2888891189 hasConcept C103278499 @default.
• W2888891189 hasConcept C115961682 @default.
• W2888891189 hasConcept C119857082 @default.
• W2888891189 hasConcept C14036430 @default.
• W2888891189 hasConcept C153180895 @default.
• W2888891189 hasConcept C154945302 @default.

• next