Why GPUs Are Essential for AI Training

Technical overview of why GPUs outperform CPUs for deep learning training: neural networks' matrix operations, parallelism, GPU architecture and GPGPU benefits.

Future Directions in Lightweight Deep Learning

Review of lightweight deep learning for resource-constrained devices: TinyML, quantization, architectures and deployment strategies for efficient inference.

Deep Learning Beginner Demo: Handwritten Digit Case

Overview of deep learning fundamentals and a TensorFlow 2 handwritten digit recognition demo, covering neurons, activation/loss functions, CNNs, training, and prediction.

Impact of Irrelevant Retrieved Passages on LLM Outputs in RAG

Retrieval-augmented generation robustness analysis: semantically related but answer-irrelevant retrieved fragments and higher fragment counts degrade LLM accuracy and confidence.

Core Principles of Major Machine Learning Models

Survey of machine learning models grouped into six categories: neural networks, symbolic, decision trees, probabilistic, nearest neighbor, and ensemble methods.

Do Larger Vision Models Always Perform Better?

Scale scaling (S2): run pretrained frozen vision models at multiple image resolutions to produce multi-scale representations, matching larger models on visual tasks.

LLM Inference Optimization for NLP Applications

System-level overview of LLM inference optimization, detailing techniques and tradeoffs to improve throughput for Transformer-based large language models.

Why GPUs Are Suited for AI: AI Server Supply Chain Analysis

Technical overview of AI servers, GPU/CPU architectures, training vs inference, compute demand and market estimates, including H100/A100 performance and China server market

In-Memory Compute Chips for ML Inference

System-level evaluation of compute-in-memory (CiM) for accelerating GEMM in ML inference: compares analog vs digital CiM, cache-level integration, and optimal dataflows.

How to Achieve a Trillion-Transistor GPU

Analysis of semiconductor advances enabling AI scale: 3D integration, CoWoS/HBM packaging, silicon photonics and energy-efficient trends toward trillion-transistor GPUs.

Five Core Network Requirements for Large AI Models

Network requirements for large-model GPU training: RDMA-based bandwidth, ultra-low latency, stability, and automated deployment for scalable multi-GPU clusters.

Advances in Deep Learning for Polarimetric Imaging

Overview of deep learning-based polarimetric imaging methods for de-scattering and denoising in complex environments, with model embedding and future directions.

Key Factors of Compute for Large AI Models

Concise technical overview of GPU concepts, architecture and GPU vs CPU differences, parallel processing and performance factors for graphics and AI inference.

ByteDance MegaScale: Techniques for 10,000+ GPU Training

MegaScale system design and deployment for efficient, stable LLM training on 10,000+ GPUs: algorithm, communication, network tuning, fault tolerance, MFU gains.

Classical Computer Vision vs. Deep Learning

Analysis of deep learning in computer vision: strengths, limits, dataset biases, comparison with classical vision methods, interpretability and risks in safety-critical applications.

Understanding Generative Adversarial Networks (GANs)

Explains how generative adversarial networks (GANs) work and shows step-by-step training of a PyTorch GAN on MNIST, including network design and training loop.

Predicting Fall Risk with Computer Vision and ML

Study summary: Azure Kinect computer vision and machine learning for gait recognition—classifying normal, pelvic obliquity, and knee hyperextension; SVM/KNN top accuracy.

Dynamic Memory Compression for KV Cache During Inference

Dynamic Memory Compression (DMC) compresses the Transformer KV cache online during autoregressive inference, improving throughput and enabling longer context windows.

Renesas RZ/V2L Pretrained AI Model

RZ/V2L MPU with DRP-AI overview and pretrained plant leaf disease classification model; runtime modes, hardware/software requirements, and inference performance.

Fourier Transform Fundamentals and Machine Learning Uses

Explains Fourier transform fundamentals, FFT use in signal processing and machine learning, and Python time-series examples for frequency-domain feature extraction.