How to improve Levenberg-Marquardt's method for polynomial curve fitting?
Some weeks ago I started coding the Levenberg-Marquardt algorithm from scratch in Matlab. I'm interested in the polynomial fitting of the data but I haven't been able to achieve the level of accuracy I would like. I'm using a fifth order polynomial after I tried other polynomials and it seemed to be the best option. The algorithm always converges to the same function minimization no matter what improvements I try to implement. So far, I have unsuccessfully added the following features:
- Geodesic acceleration term as a second order correction
- Delayed gratification for updating the damping parameter
- Gain factor to get closer to the Gauss-Newton direction or the steepest descent direction depending on the iteration.
- Central differences and forward differences for the finite difference method
I don't have experience in nonlinear least squares, so I don't know if there is a way to minimize the residual even more or if there isn't more room for improvement with this method. I attach below an image of the behavior of the polynomial for the last iterations. If I run the code for more iterations, the curve ends up not changing from iteration to iteration. As it is observed, there is a good fit from time = 0 to time = 12. But I'm not able to fix the behavior of the function from time = 12 to time = 20. Any help will be very appreciated.
Fitting a polynomial does not seem to be the best idea. Your data set looks like an exponential transient, with an horizontal asymptote. Forcing a polynomial to that will work very poorly.
I'd rather try with a simple model, such as
A (1 - e^(-at)).
With naked eye, A ~ 15. You should have a look at the values of log(15 - y).
- tensorflow:Can save best model only with val_acc available, skipping
- Keras: Difference between Kernel and Activity regularizers
- How can one run a Core ML model on macOS 10.12?
- How to fix the learning-rate for Huggingface´s Trainer?
- Roboflow Vs. Darknet for generating weight file and creating the model
- What is the difference between backpropagation and reverse-mode autodiff?
- Load Registered Component in Azure ML for Pipeline using Python sdk v2
- classifiers in scikit-learn that handle nan/null
- Encoding two categorial data present in same dataset in Deep learning
- Machine learning in q/kdb+
- Why do the sensitivity (recall) values differ between classification_report and precision_recall_fscore_support in a loop?
- How do you specify the bfloat16 mixed precision with the Intel Extension for PyTorch?
- mlflow doesn't autolog artifacts while logging images
- Choosing top k models using GridSearchCV in scikit-learn
- Calculating pairwise distances between entries in a `torch.tensor`
- Reusing a feature to split regression decision tree's nodes
- Fine tuning LayoutLmv3 using Cord-V2 dataset
- What does model.eval() do in pytorch?
- ElasticNetCV in Python: Get full grid of hyperparameters with corresponding MSE?
- Why is the accuracy for my Keras model always 0 when training?
- Polars - issues with performance - attempting to create a new dataframe per row
- Keras model.export() fails because of NoneType shapes in model
- Meta-feature analysis: split data for computation on available memory
- Single-headed transformer model, output and target tensor sizes don't match
- How to add the bounding box values to the labels text files during prediction with a trained YOLO-V8 instance segmentation model?
- How to calculate correlation of colours in a dataset?
- Convert array of indices to one-hot encoded array in NumPy
- Import custom prediction routine to vertex AI pipeline
- what is the difference between bigram and unigram text features extraction
- How to map the results of Principal Component Analysis back to the actual features that were fed into the model?