16 papers · ranked by Valyu relevance
Simone Scardapane, Elena Nieddu, Donatella Firmani, Paolo Merialdo
The design of activation functions is a growing research area in the field of neural networks. In particular, instead of using fixed pointwise functions (e.g., the rectified linear unit), several authors have proposed ways of learning these functions directly from the data in a nonparametric fashion. In this paper we…
Ahmad Navid Ghanizadeh, Kamaledin Ghiasi-Shirazi, Reza Monsefi, Mohammadreza Qaraei
'Mohammadreza Qaraei'] Multiple Kernel Learning is a conventional way to learn the kernel function in kernel-based methods. MKL algorithms enhance the performance of kernel methods. However, these methods have a lower complexity compared to deep learning models and are inferior to these models in terms of recognition…
Ara Ghukasyan, Jack S. Baker, Oktay Göktaş, Juan Carrasquilla + 1 more
'Santosh Kumar Radha'] As quantum computers become increasingly practical, so does the prospect of using quantum computation to improve upon traditional algorithms. Kernel methods in machine learning is one area where such improvements could be realized in the near future. Paired with kernel methods like support-vector…
Weiguo Gao
Length Extrapolation Authors: ['Weiguo Gao'] When the predicted sequence length exceeds the length seen during training, the transformer's inference accuracy diminishes. To address these challenges, the ability to generalize or extrapolate to longer sequence lengths has garnered increased attention. Existing relative…
Mehmet Gönen
Multiple kernel learning algorithms are proposed to combine kernels in order to obtain a better similarity measure or to integrate feature representations coming from different data sources. Most of the previous research on such methods is focused on the computational efficiency issue. However, it is still not feasible…
Nora K. Speicher, Nico Pfeifer
Personalized treatment of patients based on tissue-specific cancer subtypes has strongly increased the efficacy of the chosen therapies. Even though the amount of data measured for cancer patients has increased over the last years, most cancer subtypes are still diagnosed based on individual data sources (e.g. gene…
Jingbin Wang, Haoxiang Wang, Yi‐Hua Zhou, Nancy McDonald
—In this paper, we propose a multi-kernel classifier learning algorithm to optimize a given nonlinear and nonsmoonth multivariate classifier performance measure. Moreover, to solve the problem of kernel function selection and kernel parameter tuning, we proposed to construct an optimal kernel by weighted linear…
Nicolas Langrené, Xavier Warin
Kernel density estimation and kernel regression are powerful but computationally expensive techniques: a direct evaluation of kernel density estimates at M evaluation points given N input sample points requires a quadratic O(MN) operations, which is prohibitive for large scale problems. For this reason, approximate…
Yaqiang Yao, Yang Li, Bingbing Jiang, Huanhuan Chen
To cluster data that are not linearly separable in the original feature space, k-means clustering was extended to the kernel version. However, the performance of kernel k-means clustering largely depends on the choice of kernel function. To mitigate this problem, multiple kernel learning has been introduced into the…
Seyed Shakib Vedaie, Moslem Noori, Jaspreet S. Oberoi, Barry C. Sanders + 1 more
'Barry C. Sanders' 'Ehsan Zahedinejad'] Kernel methods play an important role in machine learning applications due to their conceptual simplicity and superior performance on numerous machine learning tasks. Expressivity of a machine learning model, referring to the ability of the model to approximate complex functions…
Hachem Kadri, Alain Rakotomamonjy, Francis Bach, Pierre‐Marie Preux
Positive definite operator-valued kernels generalize the well-known notion of reproducing kernels, and are naturally adapted to multi-output learning situations. This paper addresses the problem of learning a finite linear combination of infinite-dimensional operator-valued kernels which are suitable for extending…
Sobom M. Somé, Célestin C. Kokonendji
Associated kernels have been introduced to improve the classical continuous kernels for smoothing any functional on several kinds of supports such as bounded continuous and discrete sets. This work deals with the effects of combined associated kernels on nonparametric multiple regression functions. Using the…
Hachem Kadri, Emmanuel Duflos, Pierre‐Marie Preux, Stéphane Canu + 2 more
'Alain Rakotomamonjy' 'Julien Audiffren'] In this paper1 we consider the problems of supervised classification and regression in the case where attributes and labels are functions: a data is represented by a set of functions, and the label is also a function. We focus on the use of reproducing kernel Hilbert space…
Hachem Kadri, Asma Rabaoui, Pierre‐Marie Preux, Emmanuel Duflos + 1 more
'Alain Rakotomamonjy'] Although operator-valued kernels have recently received increasing interest in various machine learning and functional data analysis problems such as multi-task learning or functional regression, little attention has been paid to the understanding of their associated feature spaces. In this…
Pradeep Reddy Raamana
Kernel methods have proven to be powerful techniques for pattern analysis and machine learning (ML) in a variety of domains. However, many of their original or advanced implementations remain in Matlab. With the incredible rise and adoption of Python in the ML and data science world, there is a clear need for a…
Julien Audiffren, Hachem Kadri
We study the stability properties of nonlinear multi-task regression in reproducing Hilbert spaces with operator-valued kernels. Such kernels, a.k.a. multi-task kernels, are appropriate for learning problems with nonscalar outputs like multi-task learning and structured output prediction. We show that multi-task kernel…