Chuguang Pan

Last seen: 1 day ago | Active since 2019

Followers: 4 Following: 12

Message

Continuous learning matters enormously. Knowledge and industries change fast these days, so we can not stop learning once we master a set of skills. Keeping learning helps us broaden our horizons, improve our problem-solving skills and stay competitive in both work and life.

Programming Languages:
MATLAB, Python, C
Spoken Languages:
English, Chinese
Pronouns:
He/him
Professional Interests:
Control Algorithm Design, Signal and Data Processing, Mechanical Engineering

Statistics

All

View badges

Feeds

Submitted

Generalized State Space Layer for Sampling Rate Shift
Generalized State Space (GSS) layer can process input sampling rate shift based on SSM with learnable timescale.

1 day ago | 1 download |

0.0 / 5

Solved

Swap the column given with another one!
We have a random sized matrix A. You have to change the nth column with the kth column. The result must be a matrix B. For exam...

2 days ago

Question

pinv function is supported for dlarray input while the pageinv function is not supported
In the deep learning pipeline, I want to use inverse operation on every page of 3D dlarray object. I have tried pageinv function...

4 days ago | 2 answers | 0

2

answers

Submitted

Wavelet-Structured Gated Recurrent Unit Layer
Customized Wavelet-Structured Gated Recurrent Unit (WSGRU) layer for multivariate time series analysis.

4 days ago | 0 downloads |

0.0 / 5

Answered
How to multiple very large numbers in Matlab?
@Manny. You can use sym function to obtain accurate symbolic result. accurateNum = sym("111111111"); res = accurateNum.^2

4 days ago | 2

Submitted

Caputo Fractional-Order GRU for Time-Series Learning
Caputo fractional-order extension model is integrated into the GRU state update mechanism based on Caputo fractional derivative....

6 days ago | 0 downloads |

0.0 / 5

Answered
dlarray supported alternatives of matrix inverse and matrix determinant
After refering to the Example presented in Specify Custom Operation Backward Function and the Matrix calculus formula. I define ...

10 days ago | 0

Question

dlarray supported alternatives of matrix inverse and matrix determinant
I want to use matrix inverse and matrix determinant in the Dynamic Computation Graph of the deep learning model. However, there ...

10 days ago | 1 answer | 0

1

answer

Submitted

Customized MambaBlock Layer Using Minimal Implementation
Minimal implementation of MambaBlock.

12 days ago | 2 downloads |

0.0 / 5

Submitted

Flow Matching Example for Generative Modeling
FlowMatching (FW) model builds the probability path from source distribution to target distribution.

14 days ago | 4 downloads |

0.0 / 5

Solved

Find the standard deviation of non-zero elements in a matrix
Given a matrix A, calculate the sample standard deviation of only the non-zero elements. If there are fewer than 2 non-zero ...

16 days ago

Submitted

Split Input Function Layer with Multi-Outputs
Split input with uniform block size along channel ("C") dimension based on functionLayer.

18 days ago | 1 download |

0.0 / 5

Submitted

RBFKAN for Physics-Informed Machine Leaning
RBFKAN constructs its edge activation functions as a linear combination of radial basis functions.

19 days ago | 5 downloads |

0.0 / 5

Submitted

Customized Linformer Attention Layer
Linformer: multi-head linear attention with low-rank temporal projection.

22 days ago | 1 download |

0.0 / 5

Submitted

Gated Mamba Module based on Deep Learning Toolbox
Gated mamba module for multivariates time sequence deep learning.

24 days ago | 2 downloads |

0.0 / 5

Submitted

Neural Dynamics Gradient Time Acceleration Optimizer
NDGTA optimizer guides parameters of neural networks toward flat minima during the training for generalization.

25 days ago | 2 downloads |

0.0 / 5

Submitted

Deep Gaussian Process (DGP) Layer for DGP Integration
Deep Gaussian Processes (DGPs) are hierarchical Bayesian non-parametric models that extend standard Gaussian Processes for deep ...

27 days ago | 4 downloads |

0.0 / 5

Submitted

Symbolic KANs for Interpretable Learning
Symbolic Kolmogorov-Arnold Networks (Symbolic-KANs) embeds symbolic structure directly within a trainable network.

1 month ago | 3 downloads |

0.0 / 5

Submitted

Customized Spike Neuron Layers for Spiking Neural Networks
SpikeIFNeuronLayer and SpikeLIFNeuronLayer for constructing Spiking Neural Networks (SNNs).

1 month ago | 1 download |

0.0 / 5

Submitted

Tensorized LSTM Network with Mixture Attention
Tensorized LSTM unit for multivariate time series (MTS) learning. Mixture Attention for variable attention and temproal attentio...

1 month ago | 8 downloads |

0.0 / 5

Question

Custom Deep Learning Layer with different input and output format occurs error
When I construct the custom deep learning layer with formattable input and output, checkLayer function returns error when the in...

1 month ago | 1 answer | 0

1

answer

Submitted

Spiking Neural Network Example using Spiking LIF Layer
Customized Spiking LIF (Leaky-and-Firing) Layer without learnable parameters for constructing Spiking Neural Networks (SNNs).

2 months ago | 6 downloads |

0.0 / 5

Submitted

Customized Invertible Attention Layer
Invertible Attention Layer (IAL) constructed based on Deep Learning Toolbox for conditional invertible neural networks (cINNs).

2 months ago | 2 downloads |

0.0 / 5

Question

How to ensurethat the sum([0.7 0.2 0.1])==1 return true when testing equivalence relation
I find that when I want to test the equivalence between sum([0.7 0.2 0.1]) and 1, the testing result return false. I think the e...

2 months ago | 2 answers | 0

2

answers

Submitted

Wavelet Kolmogorov-Arnold Layers
Wavelet Kolmogorov-Arnold Networks (Wav-KAN) framework for enhancing interpretability and performance.

2 months ago | 14 downloads |

0.0 / 5

Submitted

Customized BiGRU Layer using Deep Learning Toolbox
BiGRU layer constructed based on Deep Learning Toolbox.

2 months ago | 3 downloads |

0.0 / 5

Question

How is the spectralConvolution1dLayer forward computation process performed
spectralConvolution1dLayer performs convolution on 1-D input using frequency domain transformations. The layer convolves the in...

2 months ago | 1 answer | 0

1

answer

Solved

MATLAB 101: Reverse a Vector
Write a MATLAB function that takes a 1D vector (either a row or a column vector) and returns the vector with its elements in exa...

3 months ago

Submitted

Multi Head Hybrid Adaptive Attention Mechanism
Multi-head hybrid adaptive attention (MHAA) mechanism integrating MSA and MEA.

3 months ago | 6 downloads |

0.0 / 5

Solved

Multiply the Diagonals of Two Vectors
Find the diagonals of vectors a and b and multiply them.

3 months ago