155 skills found · Page 3 of 6
AmeyaJagtap / Rowdy Activation FunctionsWe propose Deep Kronecker Neural Network, which is a general framework for neural networks with adaptive activation functions. In particular we proposed Rowdy activation functions that inject sinusoidal fluctuations thereby allows the optimizer to exploit more and train the network faster. Various test cases ranging from function approximation, inferring the PDE solution, and the standard deep learning benchmarks like MNIST, CIFAR-10, CIFAR-100, SVHN etc are solved to show the efficacy of the proposed activation functions.
SciML / DiffEqApproxFun.jlThe tools for proper interactions between ApproxFun.jl and DifferentialEquations.jl for pseudospectiral partial differential equation discretizations in scientific machine learning (SciML)
NaiyangGuan / Truncated Cauchy Non Negative Matrix FactorizationNon-negative matrix factorization (NMF) minimizes the euclidean distance between the data matrix and its low rank approximation, and it fails when applied to corrupted data because the loss function is sensitive to outliers. In this paper, we propose a Truncated CauchyNMF loss that handle outliers by truncating large errors, and develop a Truncated CauchyNMF to robustly learn the subspace on noisy datasets contaminated by outliers. We theoretically analyze the robustness of Truncated CauchyNMF comparing with the competing models and theoretically prove that Truncated CauchyNMF has a generalization bound which converges at a rate of order O(lnn/n‾‾‾‾‾√) , where n is the sample size. We evaluate Truncated CauchyNMF by image clustering on both simulated and real datasets. The experimental results on the datasets containing gross corruptions validate the effectiveness and robustness of Truncated CauchyNMF for learning robust subspaces.
saurabhsh96 / WFIInvScattProbCode for reconstructing the object domain contrast function using Born Approximation in inverse scattering problem
during-master-degree / M.E Energy Efficient Power And Subcarrier Allocation For OFDMA Systems With Value Function ApproximaEnergy-Efficient Power and Subcarrier Allocation for OFDMA Systems with Value Function Approximation Approach. EI paper from march to september 2012.
bfly123 / DeePolyHigh-order accuracy Neural Network framework for function approximation and PDE solving
anthony-nouy / ApproximationToolboxAn object-oriented MATLAB toolbox for the approximation of functions and tensors
tliu1997 / RNAC[NeurIPS 2023] "Natural Actor-Critic for Robust Reinforcement Learning with Function Approximation"
LaRiffle / Approximate ModelsPython Library for Function Approximation in Machine Learning
timbmg / Easy21 RlEasy21 assignment from David Silver's RL Course at UCL
stuffmatic / MicrochebyA Rust library for computing and evaluating polynomial approximations of functions of one variable using Chebyshev expansions.
rutgers-apl / Rlibm GeneratorA tool to generate approximations of elementary functions that produce the correctly rounded result for all inputs. This tool can be used for different representations that approximate real numbers.
AhmedMagdyHendawy / MINTO🌿 [ICLR 2026] Official codebase for MINTO. 🌿 MINTO is a simple, yet effective target bootstrapping method for temporal-difference RL that enables faster, more stable learning and consistently improves performance across algorithms and benchmarks.
simone-brugiapaglia / Sparse Hd BookThis is a repository associated with the book "Sparse Polynomial Approximation of High-Dimensional Functions" by Ben Adcock, Simone Brugiapaglia, and Clayton G. Webster to be published by SIAM in late 2021.
shuvoxcd01 / GridMindA library of reinforcement learning (RL) algorithms.
IgorKohan / NormalHermiteSplines.jlMultivariate Normal Hermite-Birkhoff Interpolating Splines in Julia
ntotorica / SMP Passivity EnforcementAuthor: Nathan Totorica Date: 5/14/2021 # Singularity Matrix Pertubation (SMP) This code was written for a class project in the course entitled ECE 504: "ST: Passive Electromagnetic Systems" taught in Spring 2021 by Dr. Ata Zadehgol at the University of Idaho in Moscow. This code was developed, in part, based on the code developed by Jennifer Houle in ECE 504 "ST: Modern Circuit Synthesis Algorithms" taught in Spring 2020 by Dr. Ata Zadehgol. Jennifer's code is available online at https://github.com/JenniferEHoule/Circuit_Synthesis. ## Overview Singular Matrix Perturbation (SMP) as introduced in [11], is a passivity enforcement algorithm for use on fitted models. This robust method is fast, computationally inexpensive, and accurate in enforcing passivity. This implementation using Python can easily be used with Vector Fitting algorithm implemented in [12]. Example cases to demonstrate passivity enforcmment were taken from [7][15], as well as custom examples designed in simulation software based off of multiport circuit synthesis as described in [6]. ## I/O and Flow Description - Instructions for input and output and flow description can be found in flow_diagram.pdf. ## Files - SMP.py: Singularity matrix perturbation implementation of [11] - s_compare.py: Compare ADS simulated S matrices. Calculate RMS error and plot - eig_plot.py: Plotting functions. Eigenvalue plots based off of plot.py in [12] - smp_ex.py: Example of how to pull .sp file in, parse parameters, send to vector fitting, enforce passivity, and generate new passive circuit. Imported from [12] - Ex2_y.py - vectfit3.py - intercheig.py - rot.py - pass_check.py - fitcalc.py - FRPY.py - intercheig.py - violextrema.py - plots.py - quadprog.py - pr2ss.py - utils.py ## Licensing GPL-3.0 License In addition to licensing: Embedding any of (or parts from) the routines of the Matrix Fitting Toolbox in a commercial software, or a software requiring licensing, is strictly prohibited. This applies to all routines, see Section 2.1. If the code is used in a scientific work, then reference should me made as follows: VFdriver.m and/or vectfit3.m: References [1],[2],[3] RPdriver.m and/or FRPY.m applied to Y-parameters: [8],[9] ## References [1] B. Gustavsen and A. Semlyen, "Rational approximation of frequency domain responses by Vector Fitting", IEEE Trans. Power Delivery, vol. 14, no. 3, pp. 1052-1061, July 1999. [2] B. Gustavsen, "Improving the pole relocating properties of vector fitting", IEEE Trans. Power Delivery, vol. 21, no. 3, pp. 1587-1592, July 2006. [3] D. Deschrijver, M. Mrozowski, T. Dhaene, and D. De Zutter, "Macromodeling of Multiport Systems Using a Fast Implementation of the Vector Fitting Method", IEEE Microwave and Wireless Components Letters, vol. 18, no. 6, pp. 383-385, June 2008. [4] B. Gustavsen, VFIT3, The Vector Fitting Website. March 20, 2013. Accessed on: Jan. 21, 2020. [Online]. Available: https://www.sintef.no/projectweb/vectfit/downloads/vfit3/. [5] A. Zadehgol, "A semi-analytic and cellular approach to rational system characterization through equivalent circuits", Wiley IJNM, 2015. [Online]. https://doi.org/10.1002/jnm.2119 [6] V. Avula and A. Zadehgol, "A Novel Method for Equivalent Circuit Synthesis from Frequency Response of Multi-port Networks", EMC EUR, pp. 79-84, 2016. [Online]. Available: ://WOS:000392194100012. [7] B. Gustavsen, Matrix Fitting Toolbox, The Vector Fitting Website. March 20, 2013. Accessed on: Feb. 25, 2020. [Online]. Available: https://www.sintef.no/projectweb/vectorfitting/downloads/matrix-fitting-toolbox/. [8] B. Gustavsen, "Fast passivity enforcement for S-parameter models by perturbation of residue matrix eigenvalues", IEEE Trans. Advanced Packaging, vol. 33, no. 1, pp. 257-265, Feb. 2010. [9] B. Gustavsen, "Fast Passivity Enforcement for Pole-Residue Models by Perturbation of Residue Matrix Eigenvalues", IEEE Trans. Power Delivery, vol. 23, no. 4, pp. 2278-2285, Oct. 2008. [10] A. Semlyen, B. Gustavsen, "A Half-Size Singularity Test Matrix for Fast and Reliable Passivity Assessment of Rational Models," IEEE Trans. Power Delivery, vol. 24, no. 1, pp. 345-351, Jan. 2009. [11] E. Medina, A. Ramirez, J. Morales and K. Sheshyekani, "Passivity Enforcement of FDNEs via Perturbation of Singularity Test Matrix," in IEEE Transactions on Power Delivery, vol. 35, no. 4, pp. 1648-1655, Aug. 2020, doi: 10.1109/TPWRD.2019.2949216. [12] Houle, Jennifer, GitHub. May 10, 2020. Accessed on: February 3, 2021. [Online]. Available: https://github.com/jenniferEhoule/circuit_synthesis
jjfiv / Ltr2netCode and data for "Universal Approximation Functions for Fast Learning to Rank: Replacing Expensive Regression Forests with Simple Feed-Forward Networks"
ritchie46 / FORMFirst Order Reliability Methods. Taylor series approximation of the performance function of different stochastic variables.
sachag678 / Reinforcement LearningContains baseline implementations of all RL algorithms using tabular and function approximations. Algorithms such as TD(0), MC, SARSA, Q-Learning and Policy Gradient methods.
