Add some invariant mnist results

README.md

@@ -234,6 +234,18 @@ python examples/variational_autoencoder.py -m
 ```
 
 ## Invariant MNIST (fully connected)
+Some comparisons using parameter scans maxabs normalization as default
+```
+python3 examples/invariant_mnist.py -m mlp.n=2,3,4,5,6 mlp.hidden.width=128 mlp.layer_type=polynomial optimizer=sophia
+```
+| n   | test accuracy       |
+|--------------|----------------------|
+|2 | 0.9501000046730042
+|3 | 0.9785000085830688
+|4 | 0.9711999893188477
+|5 | 0.9653000235557556
+|6 |
+
 Without polynomial refinement
 ```python
 python examples/invariant_mnist.py max_epochs=100 train_fraction=1 mlp.layer_type=continuous mlp.n=5 mlp.p_refine=False mlp.hidden.layers=4
@@ -335,11 +347,11 @@ to
 [KAN: Kolmogorov–Arnold Networks 2024](https://arxiv.org/pdf/2404.19756) was published (9 years after the original implementation of the technique in this repo), where B-splines were used on the grid. Looking at that repo, the real difference seems to be B-splines vs lagrange
 polynomials.
 
-[Variations on the Chebyshev-Lagrange Activation Function](https://arxiv.org/abs/1906.10064) implements a linear extension 
+[Variations on the Chebyshev-Lagrange Activation Function](https://arxiv.org/abs/1906.10064) implements a linear extension
 to the values beyond [-1,1] which would solve the problem of polynomial growth outside that range.
 
 [KAN: Kolmogorov–Arnold Networks: A review 2024](https://vikasdhiman.info/reviews/KAN_a_review.pdf) A review of KANs with respect to other types of networks, especially spline networks
 
-[Linear spline networks 2020](https://arxiv.org/pdf/2001.06263) 
+[Linear spline networks 2020](https://arxiv.org/pdf/2001.06263)
 
-[Learning Activation Functions in Deep (Spline) Neural Networks 2020](https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9264754) using B splines
+[Learning Activation Functions in Deep (Spline) Neural Networks 2020](https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9264754) using B splines