A compiler for a self-hosting Scheme subset produces a differentiable meta-circular interpreter supporting gradient-based co-optimization of program structure and continuous parameters.
A Differentiable Programming System to Bridge Machine Learning and Scientific Computing
12 Pith papers cite this work, alongside 130 external citations. Polarity classification is still indexing.
12
Pith papers citing it
130
external citations · Pith
abstract
Scientific computing is increasingly incorporating the advancements in machine learning and the ability to work with large amounts of data. At the same time, machine learning models are becoming increasingly sophisticated and exhibit many features often seen in scientific computing, stressing the capabilities of machine learning frameworks. Just as the disciplines of scientific computing and machine learning have shared common underlying infrastructure in the form of numerical linear algebra, we now have the opportunity to further share new computational infrastructure, and thus ideas, in the form of Differentiable Programming. We describe Zygote, a Differentiable Programming system that is able to take gradients of general program structures. We implement this system in the Julia programming language. Our system supports almost all language constructs (control flow, recursion, mutation, etc.) and compiles high-performance code without requiring any user intervention or refactoring to stage computations. This enables an expressive programming model for deep learning, but more importantly, it enables us to incorporate a large ecosystem of libraries in our models in a straightforward way. We discuss our approach to automatic differentiation, including its support for advanced techniques such as mixed-mode, complex and checkpointed differentiation, and present several examples of differentiating programs.
citation-role summary
method 2
background 1
citation-polarity summary
representative citing papers
The Neural Compiler converts symbolic programs into exact differentiable PyTorch modules for hybrid scientific machine learning, enabling precise encoding of known physics with few trainable parameters.
Derives heterogeneity bounds separating objective-shift and feasible-set-shift effects in decision-focused federated learning and shows federation benefits when statistical gains exceed client-specific penalties.
Universal Differential Equations unify scientific models with machine learning by embedding flexible approximators into differential equations, enabling applications from biological mechanism discovery to high-dimensional optimization.
VertAX supplies a differentiable JAX implementation of vertex models for confluent epithelia that enables forward simulation, mechanical parameter inference, and inverse design of tissue-scale behaviors.
Hybrid neural-process model derives biokinetic parameters from genomic traits for soil organic matter turnover, with ecological constraints, and outperforms baselines on synthetic and real data.
Standard PINNs fail to recover layer moduli in multilayer pavement systems due to domain discontinuities; XPINN improves but remains sensitive to weighting and noise; DiffFEM yields superior accuracy, stability, and efficiency on the same synthetic benchmark.
A physics-informed ML method embeds a differentiable flow simulator into neural network training to infer permeability from sparse pressure data, halving inference error versus data-driven baselines across scenarios and maintaining accuracy on extreme events.
Machine learning trains an ensemble optimal control scheme to pick optimal measurement times for non-Markovian quantum noise parameters, reaching near Cramér-Rao bound precision.
Neural Computers are introduced as a new machine form where computation, memory, and I/O are unified in a learned runtime state, with initial video-model experiments showing acquisition of basic interface primitives from traces.
This perspective paper categorizes hybrid architectures for combining mechanistic and data-driven models using residual learning, Neural ODEs, and solver-in-the-loop to model neurological disorder progression.
citing papers explorer
-
Universal Differential Equations for Scientific Machine Learning
Universal Differential Equations unify scientific models with machine learning by embedding flexible approximators into differential equations, enabling applications from biological mechanism discovery to high-dimensional optimization.
-
Learning Non-Markovian Noise via Ensemble Optimal Control
Machine learning trains an ensemble optimal control scheme to pick optimal measurement times for non-Markovian quantum noise parameters, reaching near Cramér-Rao bound precision.