منابع مشابه
Neural networks in non-Euclidean metric spaces
Multilayer Perceptrons (MLPs) use scalar products to compute weighted activation of neurons providing decision borders using combinations of soft hyperplanes. The weighted fun-in activation function corresponds to Euclidean distance functions used to compute similarities between input and weight vector. Replacing the fan-in activation function by non-Euclidean distance function offers a natural...
متن کاملMetric Derivations on Euclidean and Non-euclidean Spaces
As introduced by Weaver, (metric) derivations extend the notion of differential operators on Euclidean spaces and provide a linear differentiation theory that is well-defined on all metric spaces with Borel measures. The main result of this paper is a rigidity theorem for measures on Euclidean spaces that support a maximal number of derivations. The proof relies substantially on ideas from geom...
متن کاملEuclidean Quotients of Finite Metric Spaces
This paper is devoted to the study of quotients of finite metric spaces. The basic type of question we ask is: Given a finite metric space M and α ≥ 1, what is the largest quotient of (a subset of) M which well embeds into Hilbert space. We obtain asymptotically tight bounds for these questions, and prove that they exhibit phase transitions. We also study the analogous problem for embedings int...
متن کاملBilipschitz Embeddings of Metric Spaces into Euclidean Spaces
When does a metric space admit a bilipschitz embedding into some finite-dimensional Euclidean space? There does not seem to be a simple answer to this question. Results of Assouad [A1], [A2], [A3] do provide a simple answer if one permits some small (“snowflake”) deformations of the metric, but unfortunately these deformations immediately disrupt some basic aspects of geometry and analysis, lik...
متن کاملIsoperimetric Inequalities of Euclidean Type in Metric Spaces
1.1. Statement of the main result. The isoperimetric problem of euclidean type for a space X and given classes Ik−1, Ik, and Ik+1 of surfaces of dimension k − 1, k, and k + 1 in X , together with boundary operators Ik+1 ∂ −→ Ik ∂ −→ Ik−1 and a volume function M on each class, asks the following: Does there exist for every surface T ∈ Ik without boundary, ∂T = 0, a surface S ∈ Ik+1 with ∂S = T a...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: MATHEMATICA SCANDINAVICA
سال: 1997
ISSN: 1903-1807,0025-5521
DOI: 10.7146/math.scand.a-12620