Log-Sum-Exp Neural Networks and Posynomial Models for Convex and Log-Log-Convex Data
نویسندگان
چکیده
منابع مشابه
Twofold exp and log
This article is about twofold arithmetic [1, 2]. Here I introduce algorithms and experimental code for twofold variant of C/C++ standard functions exp() and log(), and expm1() and log1p(). Twofold function y0 + y1 ≈ f(x0 + x1) is nearly 2x-precise so can assess accuracy of standard one. Performance allows assessing on-fly: twofold texp() over double is ~10x times faster than expq() by GNU quadm...
متن کاملConvex Hull with Optimal Query Time and O ( log n · log log n ) Update Time
The dynamic maintenance of the convex hull of a set of points in the plane is one of the most important problems in computational geometry. We present a data structure supporting point insertions in amortized O(log n · log log logn) time, point deletions in amortized O(log n · log logn) time, and various queries about the convex hull in optimal O(log n) worst-case time. The data structure requi...
متن کاملInequalities for 3-log-convex Functions
This note gives a simple method for obtaining inequalities for ratios involving 3log-convex functions. As an example, an inequality for Wallis’s ratio of Gautchi-Kershaw type is obtained. Inequalities for generalized means are also considered.
متن کاملOn Isoperimetric Inequalities for Log-convex Measures
Let μ = ρdx be a Borel measure on Rd. A Borel set A ⊂ R is a solution of the isoperimetric problem if for any B ⊂ R satisfying μ(A) = μ(B) one has μ(∂A) ≤ μ(∂B), where μ(∂A) = ∫ ∂A ρ dHd−1 is the corresponding surface measure. There exists only a small number of examples where the isoperimetric problem has an exact solution. The most important case is given by Lebesgue measure λ on R, the solut...
متن کاملAccelerated Batch Learning of Convex Log-linear Models for LVCSR
This paper describes a log-linear modeling framework suitable for large-scale speech recognition tasks. We introduce modifications to our training procedure that are required for extending our previous work on log-linear models to larger tasks. We give a detailed description of the training procedure with a focus on aspects that impact computational efficiency. The performance of our approach i...
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: IEEE Transactions on Neural Networks and Learning Systems
سال: 2020
ISSN: 2162-237X,2162-2388
DOI: 10.1109/tnnls.2019.2910417