Abdullah Şamil Güser
Machine Learning
Blog
-
2025-3-23 Confidence Intervals for Regression Models : A comparison of methods of computing confidence intervals for regression models -
2025-3-14 DeepSeek : Summary documentation containing history of DeepSeek models and various URLs -
2025-3-6 Deep Dive into LLMs like ChatGPT : my takeaways from the video -
2025-3-6 DeepSeek, China, OpenAI, NVIDIA, xAI, TSMC, Stargate, and AI Megaclusters : my takeaways from the video -
2024-12-08 Advancing AI with Local Learning and Uncertainty Estimation : o1-preview summary of the video -
2024-11-24 Stanford CS229 I Machine Learning I Building Large Language Models (LLMs) : o1-preview summary of the video -
2024-05-05 CLIP - Contrastive Language-Image Pretraining : A brief explanation of CLIP method -
2023-12-22 Approximate Nearest Neighbor Methods : This blog post explores Approximate Nearest Neighbor (ANN) methods, discussing their importance, various techniques like HNSW, LSH, ANNOY, and Spill Trees, and Python libraries for implementing these methods. -
2023-12-18 Text Classification Using Class Information : How should our approach to text classfication change if our classes also have meanings. -
2023-12-13 Model Deployment Strategies : I dissect the machine learning model development lifecycle, exploring deployment strategies and techniques for effective model deployment. -
2023-08-26 Alignment Problem : Summary of the blog Musings on the Alignment Problem by Jan Leike
Paper Summaries
-
2021-10-16 Edward Hu et. al. LoRA: Low-Rank Adaptation of Large Language Models : LoRA introduces a resource-efficient approach for adapting large pre-trained language models, such as GPT-3, to specific tasks without the heavy costs of traditional fine-tuning. It maintains model quality, minimizes inference latency, and facilitates quick task-switching. -
2023-12-19 Lingling Xu et. al. Parameter-Efficient Fine-Tuning Methods for Pretrained Language Models: A Critical Review and Assessment : This paper critically reviews Parameter Efficient Fine-Tuning (PEFT) methods for large pretrained language models (PLMs), highlighting their benefits in resource-limited settings. It assesses these methods’ performance, efficiency, and memory usage across tasks like natural language understanding, machine translation, and generation. -
2023-05-17 Rohan Anil et. al. PaLM 2 Technical Report : We introduce PaLM 2, a new state-of-the-art language model that has better multilingual and reasoning capabilities and is more compute-efficient than its predecessor PaLM. -
2023-05-18 Chunting Zhou et. al. LIMA: Less Is More for Alignment : Large language models are trained in two stages: (1) unsupervised pretraining from raw text, to learn general-purpose representations, and (2) large scale instruction tuning and reinforcement learning, to better align to end tasks and user preferences. We measure the relative importance of these two stages. -
2022-03-29 Jordan Hoffmann et. al. Training Compute-Optimal Large Language Models : We investigate the optimal model size and number of tokens for training a transformer language model under a given compute budget.
Speech Processing
-
2022-08-25 MFCC : Short description for MFCC
Course Notes
- AWS Certified Machine Learning Specialty (Udemy Course Notes)
- AWS Certified Cloud Practitioner (AWS Course Notes)
Statistics
-
2023-08-25 Hypothesis Testing : Short description for hypothesis testing -
2023-08-25 z-score : Short description for z-score