Methods for Scalable Probabilistic Inference - IPAM at UCLA
Explore scalable probabilistic inference methods for astrophysics, focusing on Gaussian Processes for time series analysis and open-source tools to accelerate Bayesian inference with large datasets.
GPU Accelerated Deep Learning for Scala 3 with Storch
Discover Storch, a new PyTorch-based library for GPU-accelerated deep learning in Scala 3. Learn about tensor operations, automatic differentiation, and neural network APIs for building enterprise-grade machine learning applications.
The Fundamentals of Modern Deep Learning with PyTorch
Learn PyTorch for deep learning: train neural networks for image and text classification, explore tensor operations, automatic differentiation, and modern libraries like PyTorch Lightning for efficient GPU training.
The Most Important Algorithm in Machine Learning - Backpropagation Explained
Explore backpropagation's fundamentals, from curve fitting to gradient descent, chain rule, and computational graphs. Gain insights into this crucial machine learning algorithm's principles and applications.
Automatic Differentiation for Sparse Tensors
Explore efficient automatic differentiation for sparse tensors, leveraging novel DSLs and optimizations to outperform existing frameworks in data-intensive applications.
Abstract Interpretation for Automatic Differentiation
Explore abstract interpretation for automatic differentiation, analyzing its application in machine learning, scientific computing, and graphics.
Efficient CHAD - Optimizing Combinatory Homomorphic Automatic Differentiation
Optimizar el algoritmo CHAD para diferenciación automática en modo reverso, logrando la complejidad computacional esperada mediante técnicas funcionales y paralelismo.
Symbolic Numeric Programming in Julia
Explore Symbolics.jl for computer-algebra metaprogramming in Julia, enabling powerful code transforms, optimization, and automatic differentiation with symbolic techniques.
Inside Every Calculus Is A Little Algebra Waiting To Get Out
Explore algebraic approaches to automatic differentiation and integration in calculus, with Kotlin code examples. Learn how synthetic differential geometry simplifies complex mathematical concepts.
Teaching New Tricks to Old Programs Part 2
Explore advanced functional programming techniques for reinterpreting existing code, including hardware circuits, automatic differentiation, and interval analysis, using Haskell and GHC plugins.
Teaching New Tricks to Old Programs Part 1
Explore new interpretations for functional programs, including hardware circuits and automatic differentiation. Learn techniques for overloading lambda and organizing interpretations in Haskell.
Getting Started with Automatic Differentiation
Explore automatic differentiation in Python, its applications in optimization and computational art, using libraries like JAX, TensorFlow, and PyTorch. Gain intuition for derivatives and gradients.
Differentiable Functional Programming
Explore differentiable functional programming, covering supervised learning, gradient descent, and automatic differentiation techniques for efficient deep learning implementations in Scala.
BLASPhemy - Improving BLAS Handling in Enzyme.jl
Optimizing BLAS and Lapack routines for Enzyme.jl, improving performance and reliability in numerical computations and machine learning applications within the Julia ecosystem.
Deep Dive in Nx Backends - Lecture 1
Explore Nx Backends in Elixir, understanding their impact on performance and automatic differentiation. Learn about backend structure, runtime dispatching, and integration with Axon for deep learning.
Neural ODEs: Continuous-Time Models for Machine Learning - TMLS 2019
Explore neural networks as implicit models and their applications in scientific computing with David Duvenaud at TMLS2019.
Strategies to Integrate Data and Biogeochemical Models
Explore integrating data and biogeochemical models using Sindbad.jl framework. Learn hybrid approaches combining physics-based models with neural networks for ecosystem dynamics simulation.
ProcessSimulator.jl - A Differentiable Chemical Process Simulator
Differentiable chemical process simulator leveraging symbolic transformations and automatic differentiation for robust flowsheeting, thermodynamic property calculations, and system-level optimization in Julia.
JSOSuite.jl - One-Stop Solution for Optimization
Explore JSOSuite.jl, a comprehensive optimization package offering automatic solver selection, research-level algorithms, and flexible modeling interfaces for various problem types in Julia.
Recent Advances in EAGO.jl and its Use With JuMP.jl
Explore EAGO.jl, an open-source global optimizer for MINLPs in Julia, and its integration with JuMP for enhanced accessibility and capabilities in solving complex optimization problems.
Recent Advances in Optimization Solvers within JuliaSmoothOptimizers
Explore cutting-edge nonlinear optimization solvers in JuliaSmoothOptimizers. Learn about new packages, performance enhancements, and flexible implementations for large-scale problems using high-level linear algebra and automatic differentiation.