InTDS ArchivebySamuel Flender·Nov 12, 2024Machine Learning in Fraud Detection: A PrimerBalancing automation, accuracy, and customer experience in an ever-evolving adversarial landscapeA response icon7A response icon7
InTDS ArchivebySamuel Flender·Jul 15, 2024User Action Sequence Modeling: From Attention to Transformers and BeyondThe quest to LLM-ify recommender systems
InTDS ArchivebySamuel Flender·Apr 23, 2024LoRA: Revolutionizing Large Language Model Adaptation without Fine-TuningExploiting the low-rank nature of weight updates during fine-tuning results in orders of magnitude reduction in learnable parametersA response icon1A response icon1
InTDS ArchivebySamuel Flender·Mar 17, 2024Demystifying Mixtral of ExpertsMistral AI’s open-source Mixtral 8x7B model made a lot of waves — here’s what’s under the hoodA response icon3A response icon3
InTDS ArchivebySamuel Flender·Feb 15, 2024The Rise of Sparse Mixtures of Experts: Switch TransformersA deep-dive into the technology that paved the way for the most capable LLMs in the industry today
InTDS ArchivebySamuel Flender·Dec 10, 2023Pushing the Limits of the Two-Tower ModelWhere the assumptions behind the two-tower model architecture break — and how to go beyondA response icon3A response icon3
InTDS ArchivebySamuel Flender·Nov 14, 2023Towards Understanding the Mixtures of Experts ModelNew research reveals what happens under the hood when we train MoE models
InTDS ArchivebySamuel Flender·Oct 29, 2023The Rise of Two-Tower Models in Recommender SystemsA deep-dive into the latest technology used to debias ranking modelsA response icon5A response icon5
InTDS ArchivebySamuel Flender·Sep 29, 2023The Multi-Task Optimization ControversyDo we need special algorithms to train models on multiple tasks at the same time?A response icon1A response icon1
InTDS ArchivebySamuel Flender·Sep 5, 2023Machine Learning with Expert Models: A PrimerHow a decades-old idea enables training outrageously large neural networks todayA response icon2A response icon2
