Machine Learning on grzenowhttp://www.grzenow.com/categories/machine-learning/Recent content in Machine Learning on grzenowHugoenWed, 15 Jan 2025 00:00:00 +0000Deep Neural Networks projectshttp://www.grzenow.com/projects/deep-neural-networks/Wed, 15 Jan 2025 00:00:00 +0000http://www.grzenow.com/projects/deep-neural-networks/<p>This is a collection of projects completed for the Deep Neural Networks course, including:</p> <ol> <li><strong>Multilayer Perceptrons</strong>: <ul> <li>Implementation of <em>ProxyProp</em> (Proximal Backpropagation) algorithm described in this <a href="https://arxiv.org/pdf/1706.04638v3" class="external-link" target="_blank" rel="noopener">paper</a>.</li> </ul> </li> <li><strong>Computer Vision</strong>: <ul> <li>Implementation of several transformations, <a href="https://arxiv.org/pdf/1911.09665" class="external-link" target="_blank" rel="noopener"><em>AdvProp</em></a>, and <a href="https://openreview.net/pdf?id=QzcZb3fWmW" class="external-link" target="_blank" rel="noopener"><em>SparseTopK</em></a> methods.</li> </ul> </li> <li><strong>Natural Language Processing</strong>: <ul> <li>Implementation of <a href="https://arxiv.org/pdf/2002.05202v1" class="external-link" target="_blank" rel="noopener"><em>SwiGLU</em></a> and <a href="https://arxiv.org/pdf/2410.05258" class="external-link" target="_blank" rel="noopener"><em>Differential Attention</em></a> for a decoder-only Transformer architecture.</li> </ul> </li> <li><strong>Reinforcement Learning</strong>: <ul> <li>Implementation of <em>Proximal Policy Optimization</em> (PPO) and <em>Random Network Distillation</em> (RND) for <a href="https://github.com/facebookresearch/minihack" class="external-link" target="_blank" rel="noopener">MiniHack</a> environments.</li> </ul> </li> </ol> <p><a href="https://github.com/grzenow4/DNN" class="external-link" target="_blank" rel="noopener">Source code</a></p>