David Bressler·Mar 24, 2025Top 10 Tips for Cutting Costs in ML SystemsBuilding out an ML product often feels like a whirlwind of experiments, training jobs, and quick iterations. Before you know it, you’re…
David Bressler·Mar 6, 2025How to Use ChatGPT to Format Math for MediumI recently published a math-heavy article on Medium through the Data Science Collective:
InData Science CollectivebyDavid Bressler·Feb 25, 2025The Elegant Symmetry of Forward and Backward Passes in Residual NetworksThis post describes a delightful ‘aha’ moment I experienced, when I figured out that the backward pass of information in a residual neural…A response icon1A response icon1
InData Science CollectivebyDavid Bressler·Feb 18, 2025How Transformers Evolved: The Three Breakthroughs That Made Transformers Scalable and PowerfulPerhaps the most surprising aspect of the evolution of the Transformer architecture for language modeling is how consistent it has remained…
InTDS ArchivebyDavid Bressler·Dec 11, 2018Building a convolutional neural network for natural language processingHow to build a gated convolutional neural network (GCNN) from scratch, with Pytorch implementationA response icon4A response icon4
InTDS ArchivebyDavid Bressler·Nov 14, 2018Speed up your deep learning language model up to 1000% with the adaptive softmax, Part 2: Pytorch…In Part 1 of this blog post, I explained how the adaptive softmax works, and how it can speed up your language model by up to 1000%. Here…A response icon2A response icon2
InTDS ArchivebyDavid Bressler·Nov 14, 2018Speed up your deep learning language model up to 1000% with the adaptive softmax, Part 1How would you like to speed up your language modeling (LM) tasks by 1000%, with nearly no drop in accuracy? A recent paper from Grave et…A response icon1A response icon1