mlsanityData to 14 Oct 2026

Automatic differentiation in ML: Where we are and where we should be going

arXiv 1810.11530 ↗First posted 26 Oct 2018cs.LG cs.PL stat.MLcited by 20

Not on the map. Why?

Abstract (arXiv)

We review the current state of automatic differentiation (AD) for array programming in machine learning (ML), including the different approaches such as operator overloading (OO) and source transformation (ST) used for AD, graph-based intermediate representations for programs, and source languages. Based on these insights, we introduce a new graph-based intermediate representation (IR) which specifically aims to efficiently support fully-general AD for array programming. Unlike existing dataflow programming representations in ML frameworks, our IR naturally supports function calls, higher-order functions and recursion, making ML models easier to implement. The ability to represent closures allows us to perform AD using ST without a tape, making the resulting derivative (adjoint) program amenable to ahead-of-time optimization using tools from functional language compilers, and enabling higher-order derivatives. Lastly, we introduce a proof of concept compiler toolchain called Myia which uses a subset of Python as a front end.

Key facts. Cited by 20 papers in the map (data through 14 Oct 2026). First posted on arXiv 26 Oct 2018.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 6; 2022: 7; 2023: 4; 2024: 3.

20 papers.

Sort:

  1. 2021MixMo: Mixing Multiple Inputs for Multiple Outputs via Deep Subnetworkscited by 32
  2. 2022Monte Carlo PINNs: deep learning approach for forward and inverse problems involving high dimensional fractional partial differential equationscited by 20
  3. 2024Physics-Informed Geometry-Aware Neural Operatorcited by 19
  4. 2022On learning agent-based models from datacited by 14
  5. 2021Applications of the Free Energy Principle to Machine Learning and Neurosciencecited by 9
  6. 2022Physics-informed Neural Networks with Periodic Activation Functions for Solute Transport in Heterogeneous Porous Mediacited by 8
  7. 2023Generalized Neural Closure Models with Interpretabilitycited by 8
  8. 2022Selected Trends in Artificial Intelligence for Space Applicationscited by 5
  9. 2024Solving the Discretised Multiphase Flow Equations with Interface Capturing on Structured Grids Using Machine Learning Librariescited by 5
  10. 2021A parameter refinement method for Ptychography based on Deep Learning conceptscited by 4
  11. 2022r-Adaptive Deep Learning Method for Solving Partial Differential Equationscited by 3
  12. 2022Machines of finite depth: towards a formalization of neural networkscited by 2
8 more
  1. 2021SelFSR: Self-Conditioned Face Super-Resolution in the Wild via Flow Field Degradation Networkcited by 1
  2. 2023Forecasting the 2016-2017 Central Apennines Earthquake Sequence with a Neural Point Processcited by 1
  3. 2023Auto-Differentiation of Relational Computations for Very Large Scale Machine Learningcited by 1
  4. 2021Coarsening Optimization for Differentiable Programmingcited by 0
  5. 2021A Brief Introduction to Automatic Differentiation for Machine Learningcited by 0
  6. 2022Mining Explainable Predictive Features for Water Quality Managementcited by 0
  7. 2023Accurate and Fast Fischer-Tropsch Reaction Microkinetics using PINNscited by 0
  8. 2024Transformed Physics-Informed Neural Networks for The Convection-Diffusion Equationcited by 0