Homework #0

EE 641: Fall 2025

Assignment Details

Assigned: 27 August
Due: Tuesday, 02 September at 23:59

BrightSpace Assignment: Homework 0

Problem 1: Residual Network for CIFAR-100

Construct a residual MLP for image classification on the CIFAR-100 dataset (https://www.cs.toronto.edu/~kriz/cifar.html) using PyTorch. Train your model to at least 68% accuracy.

Review the original ResNet paper:

K. He, X. Zhang, S. Ren and J. Sun, Deep Residual Learning for Image Recognition, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770-778.

Implement the residual network described in the table below. Use a block-architecture to define the convX layers. Combine blocks to construct the end-to-end network. Experiment with different regularization, optimizers, and learning rate schedules and apply data pre-processing and augmentations to improve model performance and generalization.

Layer Name	Architecture
conv1	3×3, 32, stride 1, padding 1 BatchNorm, ReLU, Dropout
conv2_x	$\begin{bmatrix} 3 \times 3, 32 \\ 3 \times 3, 32 \end{bmatrix} \times 2$, stride 1, padding 1
conv3_x	$\begin{bmatrix} 3 \times 3, 64 \\ 3 \times 3, 64 \end{bmatrix} \times 4$, stride 2, padding 1
conv4_x	$\begin{bmatrix} 3 \times 3, 128 \\ 3 \times 3, 128 \end{bmatrix} \times 4$, stride 2, padding 1
conv5_x	$\begin{bmatrix} 3 \times 3, 256 \\ 3 \times 3, 256 \end{bmatrix} \times 2$, stride 2, padding 1
output	max-pool, 100-d fc, softmax

Table: Residual architecture. Building blocks are shown in brackets with the numbers of blocks stacked. Downsampling is performed by conv3_x, conv4_x, and conv5_x with a stride of 2.

Use a validation set for hyperparameter tuning (e.g., grid search, random search). Record final loss and accuracy values for each hyperparameter configuration that you evaluate. Plot the training and validation classification accuracy as a function of epoch on separate figures for at least three hyperparameter configurations. Apply your best trained model to the test set and report the final loss and accuracy.

You may refer to the PyTorch ResNet reference implementation but you must implement the model from scratch:
https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py

What to submit

Tested hyperparameter configurations with final loss and accuracy values.
Accuracy and loss plots for at least three hyperparameter configurations.
Determine and report the output sizes for each convX layer.

Submission Instructions

Problem 1: ResNet MLP

Submit a summary of your network architecture and methods as well as required figures and answers/analysis. Include Python file(s) as appendix. A suitably annotated Jupyter notebook with inline analysis is sufficient.

Do NOT submit models or coefficient files.

GradeScope: Problem 1

--- title: "Homework #0" subtitle: "EE 641: Fall 2025" format: html: toc: false number-sections: false code-fold: show code-tools: true theme: cosmo --- ::: {.callout-important} ## Assignment Details **Assigned:** 27 August **Due:** Tuesday, 02 September at 23:59 **BrightSpace Assignment:** [Homework 0](https://brightspace.usc.edu) ::: --- ## Problem 1: Residual Network for CIFAR-100 Construct a residual MLP for image classification on the CIFAR-100 dataset (<https://www.cs.toronto.edu/~kriz/cifar.html>) using PyTorch. Train your model to at least 68% accuracy. Review the original ResNet paper: > K. He, X. Zhang, S. Ren and J. Sun, *Deep Residual Learning for Image Recognition*, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770-778. Implement the residual network described in the table below. Use a *block*-architecture to define the convX layers. Combine blocks to construct the end-to-end network. Experiment with different regularization, optimizers, and learning rate schedules and apply data pre-processing and augmentations to improve model performance and generalization. | Layer Name | Architecture | |------------|--------------| | conv1 | 3×3, 32, stride 1, padding 1<br>BatchNorm, ReLU, Dropout | | conv2_x | $\begin{bmatrix} 3 \times 3, 32 \\ 3 \times 3, 32 \end{bmatrix} \times 2$, stride 1, padding 1 | | conv3_x | $\begin{bmatrix} 3 \times 3, 64 \\ 3 \times 3, 64 \end{bmatrix} \times 4$, stride 2, padding 1 | | conv4_x | $\begin{bmatrix} 3 \times 3, 128 \\ 3 \times 3, 128 \end{bmatrix} \times 4$, stride 2, padding 1 | | conv5_x | $\begin{bmatrix} 3 \times 3, 256 \\ 3 \times 3, 256 \end{bmatrix} \times 2$, stride 2, padding 1 | | output | max-pool, 100-d fc, softmax | **Table:** Residual architecture. Building blocks are shown in brackets with the numbers of blocks stacked. Downsampling is performed by conv3_x, conv4_x, and conv5_x with a stride of 2. Use a validation set for hyperparameter tuning (e.g., grid search, random search). Record final loss and accuracy values for each hyperparameter configuration that you evaluate. Plot the training and validation classification accuracy as a function of epoch on separate figures for at least three hyperparameter configurations. Apply your best trained model to the test set and report the final loss and accuracy. You may refer to the PyTorch ResNet reference implementation but you must implement the model *from scratch*: <https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py> ### What to submit a. Tested hyperparameter configurations with final loss and accuracy values. b. Accuracy and loss plots for at least three hyperparameter configurations. c. Determine and report the output sizes for each `convX` layer. --- ::: {.callout-tip} ## Submission Instructions ### Problem 1: ResNet MLP Submit a summary of your network architecture and methods as well as required figures and answers/analysis. Include Python file(s) as appendix. A suitably annotated Jupyter notebook with inline analysis is sufficient. **Do NOT submit models or coefficient files.** **GradeScope:** [Problem 1](https://gradescope.com) :::