Learning computationally efficient dictionaries and their implementation as fast transforms
نویسندگان
چکیده
Dictionary learning is a branch of signal processing and machine learning that aims at finding a frame (called dictionary) in which some training data admits a sparse representation. The sparser the representation, the better the dictionary. The resulting dictionary is in general a dense matrix, and its manipulation can be computationally costly both at the learning stage and later in the usage of this dictionary, for tasks such as sparse coding. Dictionary learning is thus limited to relatively small-scale problems. In this paper, inspired by usual fast transforms, we consider a general dictionary structure that allows cheaper manipulation, and propose an algorithm to learn such dictionaries –and their fast implementation– over training data. The approach is demonstrated experimentally with the factorization of the Hadamard matrix and with synthetic dictionary learning experiments.
منابع مشابه
An Improved Particle Swarm Optimizer Based on a Novel Class of Fast and Efficient Learning Factors Strategies
The particle swarm optimizer (PSO) is a population-based metaheuristic optimization method that can be applied to a wide range of problems but it has the drawbacks like it easily falls into local optima and suffers from slow convergence in the later stages. In order to solve these problems, improved PSO (IPSO) variants, have been proposed. To bring about a balance between the exploration and ex...
متن کاملFast Continuous Haar and Fourier Transforms of Rectilinear Polygons from VLSI Layouts
We develop the pruned continuous Haar transform and the fast continuous Fourier series, two fast and efficient algorithms for rectilinear polygons. Rectilinear polygons are used in VLSI processes to describe design and mask layouts of integrated circuits. The Fourier representation is at the heart of many of these processes and the Haar transform is expected to play a major role in techniques e...
متن کاملLearning Overcomplete Signal Sparsifying Transforms
I. TRANSFORM LEARNING The formulations for learning synthesis [1] and analysis [2], [3] sparsifying dictionaries are typically non-convex and NP-hard, and the approximate algorithms are still computationally expensive. As an alternative, we recently introduced an approach for learning square sparsifying transforms W ∈ Rm×n, m = n [4], which are competitive with overcomplete synthesis or analysi...
متن کاملCalculation of One-dimensional Forward Modelling of Helicopter-borne Electromagnetic Data and a Sensitivity Matrix Using Fast Hankel Transforms
The helicopter-borne electromagnetic (HEM) frequency-domain exploration method is an airborne electromagnetic (AEM) technique that is widely used for vast and rough areas for resistivity imaging. The vast amount of digitized data flowing from the HEM method requires an efficient and accurate inversion algorithm. Generally, the inverse modelling of HEM data in the first step requires a precise a...
متن کاملDictionary Learning in Stereo Imaging
This paper presents a new method for learning overcomplete dictionaries adapted to efficient joint representation of stereo images. We first formulate a sparse stereo image model where the multi-view correlation is described by local geometric transforms of dictionary atoms in two stereo views. A maximum-likelihood method for learning stereo dictionaries is then proposed, which includes a multi...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- CoRR
دوره abs/1406.5388 شماره
صفحات -
تاریخ انتشار 2014