Deep Learning

concepts · 67 notes linked

Related: Neural Networks · Pytorch · Google · Tensorflow · Machine Learning · Yann LeCun · Reinforcement Learning · Computer Vision

Notes

• 35 Best Resources To Learn Machine Learning — Curated list of free interactive ML and deep learning visualization tools
• A New Link to an Old Model Could Crack the Mystery of Deep Learning | Quanta Magazine — Infinite-width neural networks mathematically equivalent to kernel machines
• A New Study by Google and DeepMind Introduces Geometric Complexity (GC) for Neural Network Analysis and Understanding of Deep Learning Models — Geometric Complexity measure explains regularization and double-descent in neural networks
• A Short Chronology Of Deep Learning For Tabular Data — Chronological survey comparing deep learning versus gradient boosting on tabular data
• A Tour of Machine Learning Algorithms - MachineLearningMastery.com — Taxonomy of machine learning algorithms by learning style and similarity
• A detailed example of how to generate your data in parallel with PyTorch Star — Using PyTorch Dataset and DataLoader with multicore parallel data generation
• AI Debate 2: Night of a thousand AI scholars — Sixteen scholars debate moving AI beyond deep learning
• AI Designs Computer Chips We Can't Understand — But They Work Really Well — AI inverse design produces unintuitive but high-performance RF chips
• AI critic Gary Marcus: Meta's LeCun is finally coming around to the things I said years ago — Gary Marcus argues deep learning alone cannot achieve general intelligence
• AI's Strange Chip Designs Are Faster, Smarter, and Game-Changing — AI generates unintuitive wireless chip designs that outperform human-crafted ones
• Basic Linear Algebra for Deep Learning and Machine Learning Python Tutorial — Introductory linear algebra tutorial for ML and deep learning with Python
• BearID: Face recognition for brown bears - Raspberry Pi — Deep learning bear face recognition for wildlife conservation monitoring
• Building a Simple Python-Based GAN in 5 minutes — Minimal PyTorch GAN with generator-discriminator adversarial training loop
• Building a simple Generative Adversarial Network (GAN) using TensorFlow | DigitalOcean — TensorFlow GAN tutorial with generator, discriminator, and feature visualization
• CS285 — UC Berkeley graduate deep reinforcement learning course lecture index
• Cheatsheet — Comprehensive reference index of Torch7 packages, tutorials, and GPU setup
• Cheatsheet — Torch7 Lua deep learning framework comprehensive reference guide
• ConvNetJS — JavaScript library for training neural networks in the browser
• Data Science Most Read Articles — Curated quarterly top-read data science newsletter articles from 2013-2014
• Deep Learning — Goodfellow, Bengio, Courville canonical deep learning textbook freely online
• Deep Learning Through the Lens of Example Difficulty — Prediction depth as per-example measure of deep learning difficulty
• DeepDream | TensorFlow Core — Visualizing neural network patterns via gradient ascent on images
• Facebook's Quest to Build an Artificial Brain Depends on This Guy — Yann LeCun pioneers convolutional neural networks and leads Facebook AI lab
• Generative Adversarial Networks (d2l.ai) — GAN architecture as minimax game between generator and discriminator networks
• Getting Started With Pytorch In Google Collab With Free GPU — Beginner tutorial on PyTorch tensors and autograd using free Colab GPU
• Getting an all-optical AI to handle non-linear math — Photonic chip runs neural network at 410 picosecond latency
• GitHub - catalyst-team/dl-course: Deep Learning with Catalyst — Open-source deep learning course using PyTorch and Catalyst framework
• GitHub - codecrafters-io/build-your-own-x: Master programming by recreating your favorite technologies from scratch. — Curated collection of step-by-step guides for building technologies from scratch
• GitHub - eriklindernoren/ML-From-Scratch: Machine Learning From Scratch. Bare bones NumPy implementations of machine learning models and algorithms with a focus on accessibility. Aims to cover everything from linear regression to deep learning. — Bare-bones NumPy educational implementations of ML algorithms
• GitHub - google-research/tuning_playbook: A playbook for systematically maximizing the performance of deep learning models. — Systematic hyperparameter tuning playbook from Google Brain researchers
• GitHub - pkmital/tensorflow_tutorials: From the basics to slightly more interesting applications of Tensorflow — TensorFlow tutorial series from basics through autoencoders and residual networks
• GitHub - sktime/pytorch-forecasting: Time series forecasting with PyTorch — High-level PyTorch library for deep learning time series forecasting
• GitHub - tensorflow/skflow: Simplified interface for TensorFlow (mimicking Scikit Learn) for Deep Learning — SkFlow simplified TensorFlow API mimicking scikit-learn, merged into TensorFlow contrib
• Gradient Descent Models Are Kernel Machines (Deep Learning) — Deep networks trained by gradient descent are kernel machines
• How Computationally Complex Is a Single Neuron? | Quanta Magazine — Single biological neuron equivalent to 5–8 layer deep neural network
• How To Install TensorFlow on M1 Mac — Step-by-step TensorFlow setup on Apple M1 ARM64 chip
• Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels — Data augmentation technique enabling model-free RL directly from pixels
• ImageNet — Large-scale image database organized by WordNet hierarchy for CV research
• Is AI a danger to humanity or our salvation? — Hinton, LeCun, and Bengio split on AI existential risk after ChatGPT
• Keras inventor Chollet charts a new direction for AI: a Q&A — Chollet critiques deep learning limits and proposes ARC generalization benchmark
• Learn PyTorch for Deep Learning – Free 26-Hour Course — Free 26-hour hands-on PyTorch deep learning course for beginners
• Machine-Learning Maestro Michael Jordan on the Delusions of Big Data and Other Huge Engineering Efforts — Michael Jordan critiques deep learning hype and big data statistical risks
• Neural Networks and Deep Learning — Chapter 4 — Visual proof that neural networks with hidden layers can approximate any continuous function
• Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems — Survey of offline RL algorithms learning policies from static datasets without online interaction
• Paths to the Future: A Year at Google Brain — Personal account of working at Google Brain on TensorFlow development
• PirateNets: Physics-informed Deep Learning with Residual Adaptive Networks — Adaptive residual architecture fixing deep PINN training instability
• PyTorch Explained: From Automatic Differentiation to Training Custom Neural Networks | Towards Data Science — PyTorch core abstractions from tensors and autograd to transformer encoder
• PyTorch-GAN vanilla GAN implementation — Reference PyTorch implementation of vanilla GAN on MNIST
• Quoc Le's Lectures on Deep Learning — Gaurav Trivedi — Content guide for Quoc Le's 3-lecture deep learning series at MLSS 2014
• Rainbow: Combining Improvements in Deep Reinforcement Learning — Empirical combination of six DQN extensions achieving state-of-the-art Atari performance
• Raman spectroscopy in open world learning settings using the Objectosphere approach — Objectosphere loss reduces false positives for unknown Raman spectra classes
• Running PyTorch on the M1 GPU — Benchmark of PyTorch M1 GPU support via MPS backend for deep learning training
• Salesforce AI Research Proposes 'DeepTime,' A Deep Time-Index Based Model Trained Via A Meta-Learning Formulation To Automatically Learn A Representation Function From Time-Series Data — Deep time-index meta-learning model for non-stationary time-series forecasting
• Self-supervised learning: The plan to make deep learning data-efficient - TechTalks — Yann LeCun's AAAI 2020 roadmap for data-efficient self-supervised learning
• Sparse Models, The Math, And A New Theory For GroundBreaking AI — Poggio's theory that compositional sparsity explains deep network effectiveness
• Study: Deep neural networks don't see the world the way we do — MIT study reveals deep neural networks build idiosyncratic invariances unlike human perception
• Swift: Google's bet on differentiable programming — Google's Swift for TensorFlow project integrating native differentiable programming
• The 'Godfather of AI' Has a Hopeful Plan for Keeping Future AI Friendly — Geoffrey Hinton's views on LLM risks and analog computing as AI safety mitigation
• The Doomsday Invention — Nick Bostrom's superintelligence thesis and AI existential risk debate
• The Matrix Calculus You Need For Deep Learning — Matrix calculus tutorial covering gradients and Jacobians for neural network training
• Using AI to push the boundaries of wildlife survey technologies — Deep learning counts 500,000 wildebeest from satellite imagery automatically
• Using Deep Learning to Inform Differential Diagnoses of Skin Diseases — Deep learning system matches dermatologist accuracy on 26 skin conditions
• VanillaNet — A Novel Neural Network Architecture for Computer Vision — Simple convolutional network without shortcuts or attention matches complex model performance
• Welcome to the UvA Deep Learning Tutorials! — University of Amsterdam Jupyter notebook deep learning course covering PyTorch and JAX
• What Really Made Geoffrey Hinton Into an AI Doomer — Hinton's reasons for leaving Google to warn about accelerating AI risk
• Why does Deep Learning work? — Deep learning energy landscape as spin funnel not spin glass
• ccv — Open-source C vision library releasing near-state-of-the-art CNN image classifier