1
$\begingroup$

Problem discription

I have a forth order polynomial function clearly defined. I would sample $n$ number of points and feed them to my function to get the corresponding $y$ values. I then apply a FC neural network to learn this function, which I presume should be quite a simple task.

When I perform random sampling from the input space, the neural network learns just fine. I can see the training loss getting smaller and in just a few epochs, gets pretty close to 0. Now, I change my sampling method to a special sampling method that I have, and repeat the same process. Suddenly the NN seems to not learn the function at all. And I don't mean it's generalizing poorly at the test set, it's not fitting to the training set. The loss training loss stays stagnant every epoch.

This is extremely perplexing to me. The NN learns just fine when the inputs are sampled randomly, but fails completely when another set of inputs which I generated using alternative method was used. When I examine the predicted values from the model, it seems it's predicting everything with just the same output value. This was not the case when it trained on random generated input.

Properties of the problematic data

I tried to do a tSNE on my problematic input data, and color them by y value. The low and high values seems to seperate themselves pretty well in this space, so I don't see why the NN just can't learn it properly.

tSNE plot

Per Davidmh's suggestion, I looked at a histogram of the y value distribution. Horizontal axis being the function (y) value, vertical axis is count. On the left is the function value of thee problematic input, right is for random input.

histogram

Code

The parameters of the polynomial and the problematic inputs can be found here: https://drive.google.com/drive/folders/14xqPK7M8msCJpdJ6qsHuLggNt90gcPjC

  1. Original function that I sampled from. Input is 32 dimensions
import numpy as np
from sklearn.preprocessing import PolynomialFeatures
from sklearn.linear_model import LinearRegression

# 4th degree polynomial features
poly = PolynomialFeatures(4)

# regression parameters stored in json file
reg = LinearRegression()
with open('regression.txt', 'r') as file:
    reg_params = json.load(file)
reg.coef_ = np.array(reg_params['coef_'])
reg.intercept_ = reg_params['intercept_']
reg.rank_ = reg_params['rank_']
reg.singular_ = np.array(reg_params['singular_'])

def source_function(X):
    try:
        X = poly.transform(X)
        return reg.predict(X)
    except ValueError:
        X = poly.transform(X.reshape(1,-1))
        return reg.predict(X)[0]
  1. Neural network
import torch
import torch.nn as nn
import torch.optim as optim
import torch.utils.data as Data
from torch.nn import Linear, ReLU

# MLP class
class mlp(nn.Module):
  def __init__(self, **kwargs):
    super().__init__()
    self.layer1 = nn.Linear(in_features=kwargs["input_shape"], out_features=48)
    self.layer2 = nn.Linear(in_features=48, out_features=48)
    self.layer3 = nn.Linear(in_features=48, out_features=48)
    self.layer4 = nn.Linear(in_features=48, out_features=48)
    self.pred = nn.Linear(in_features=48, out_features=1)
  
  def forward(self, features):
    x = self.layer1(features)
    x = torch.relu(x)
    x = self.layer2(x)
    x = torch.relu(x)
    x = self.layer3(x)
    x = torch.relu(x)
    x = self.layer4(x)
    x = torch.relu(x)
    prediction = self.pred(x)
    return prediction
  1. Sampled data
# load up datapoints, this is the problematic dataset
# input domain is limited to [-1, 1] for all dimensions

X_train = np.genfromtxt('X_train.csv', delimiter=',')
# get corresponding y values by feeding X_train to defined function
y_train = source_function(X_train)

# a randomly generated dataset
X_train_random = scipy.sparse.random(10000, 32, density=0.25).A
X_train_random  = X_train_random *2-1 # inputs to be within [-1, 1]
y_train_random = source_function(X_train_random)
  1. Training code
train_dataset = Data.TensorDataset(torch.from_numpy(X_train).type(torch.DoubleTensor),
                                   torch.from_numpy(y_train).type(torch.DoubleTensor))

train_loader = Data.DataLoader(dataset=train_dataset, batch_size=32, 
                               shuffle=True, num_workers=2,)

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

model = mlp(input_shape=dim).to(device)
model = model.double()

optimizer = optim.Adam(model.parameters(), lr=5e-3)

criterion = nn.MSELoss()

# Training loop

epochs = 500
for epoch in range(epochs):
    loss = 0
    for batch_features, out in train_loader:
        batch_features = batch_features.to(device)
        
        optimizer.zero_grad()
        
        outputs = model(batch_features.double())
        
        train_loss = criterion(outputs, out)
        train_loss.backward()
        
        optimizer.step()

        loss += train_loss.item()
    
    # compute the epoch training loss
    loss = loss / len(train_loader)
    
    # display the epoch training loss
    print("epoch : {}/{}, loss = {:.6f}".format(epoch + 1, epochs, loss))

Hardware: Google Colab non-GPU

Any help would be greatly appreciated!

$\endgroup$
14
  • 2
    $\begingroup$ I edited your question to remove the link. Pickles are generally considered as unsafe format, opening untrusted pickles can launch arbitrary code on your computer. If you want to share the data and code, please use formats that are auditable (e.g. source code, data in csv files etc). $\endgroup$
    – Tim
    Sep 10 at 5:06
  • $\begingroup$ Can you describe your special sampling procedure? How does it work? What are its properties? $\endgroup$
    – Sycorax
    Sep 10 at 5:08
  • $\begingroup$ Hi @Tim, thanks for the warning, I didn't know that. What format do you suggest me to use? $\endgroup$ Sep 10 at 5:08
  • $\begingroup$ Use csv, JSON or any other format that stores the data in plaintext. $\endgroup$
    – Tim
    Sep 10 at 5:11
  • $\begingroup$ Hi @Sycorax, I have a stochastic population based optimizer, which generates a bunch of candidate points per iteration. I record down all the candidate points generated in this whole process as I try to find the minima of this function. To prevent a large proportion of my points from coming from the minima region, I run the same optimization process on the maxima as well. $\endgroup$ Sep 10 at 5:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.