arXiv subjects

Roy S. Freedman

Publications and source records attributed to Roy S. Freedman.

5 recordsLinked to original sources

arXiv2024

N-Way Joint Mutual Exclusion Does Not Imply Any Pairwise Mutual Exclusion for Propositions

Given a set of N propositions, if any pair is mutual exclusive, then the set of all propositions are N-way jointly mutually exclusive. This paper provides a new general counterexample to the converse. We prove that for any set of N propositional variables, there exist N propositions such that their N-way conjunction is zero, yet all k-way component conjunctions are non-zero. The consequence is that N-way joint mutual exclusion does not imply any pairwise mutual exclusion. A similar result is true for sets since propositional calculus and set theory are models for two-element Boolean algebra.

cs.DM↗

arXiv2023

Copula Representations and Error Surface Projections for the Exclusive Or Problem

The exclusive or (xor) function is one of the simplest examples that illustrate why nonlinear feedforward networks are superior to linear regression for machine learning applications. We review the xor representation and approximation problems and discuss their solutions in terms of probabilistic logic and associative copula functions. After briefly reviewing the specification of feedforward networks, we compare the dynamics of learned error surfaces with different activation functions such as RELU and tanh through a set of colorful three-dimensional charts. The copula representations extend xor from Boolean to real values, thereby providing a convenient way to demonstrate the concept of cross-validation on in-sample and out-sample data sets. Our approach is pedagogical and is meant to be a machine learning prolegomenon.

cs.LG↗

arXiv2019

Operational Interpretations of the Chernoff Inequality

We utilize operational methods to generalize the Chernoff inequality and prove a new result that relates the moment bound to strictly absolute monotonic functions. We show that the Chernoff bound is part of a continuum of probability bounds.

math.PR↗

arXiv2019

Visual Backpropagation

We show how a declarative functional programming specification of backpropagation yields a visual and transparent implementation within spreadsheets. We call our method Visual Backpropagation. This backpropagation implementation exploits array worksheet formulas, manual calculation, and has a sequential order of computation similar to the processing of a systolic array. The implementation uses no hidden macros nor user-defined functions; there are no loops, assignment statements, or links to any procedural programs written in conventional languages. As an illustration, we compare a Visual Backpropagation solution to a Tensorflow (Python) solution on a standard regression problem.

cs.LG↗

arXiv2015

Some New Results on Binary Relations

It is well known that if a function from set A to set B has a right inverse then the function is a surjection and the right inverse is an injection. For finite sets, the number of functions, injections, and surjections can also be counted. Relations generalize functions: do similar results exist for relations? This paper proves several new results concerning binary relations. For finite sets, we derive formulas for the number of right total, right unique, left total, and left unique relations. We also provide formulas that count the number of relations that are both right unique and left unique; right unique and right total; and left unique and left total. We conclude by discussing the probability that a relation selected at random is right unique or right total.

cs.DM↗