# The Use of Deep Learning for Symbolic Integration: A Review of (Lample and Charton, 2019)

@article{Davis2019TheUO, title={The Use of Deep Learning for Symbolic Integration: A Review of (Lample and Charton, 2019)}, author={Ernest Davis}, journal={ArXiv}, year={2019}, volume={abs/1912.05752} }

Lample and Charton (2019) describe a system that uses deep learning technology to compute symbolic, indefinite integrals, and to find symbolic solutions to first- and second-order ordinary differential equations, when the solutions are elementary functions. They found that, over a particular test set, the system could find solutions more successfully than sophisticated packages for symbolic mathematics such as Mathematica run with a long time-out. This is an impressive accomplishment, as far as… Expand

#### Supplemental Code

Github Repo

Via Papers with Code

This repo will be used to collect my visualizations of press accounts of machine learning developments

#### Topics from this paper

#### 12 Citations

A Neuro-Symbolic Method for Solving Differential and Functional Equations

- Computer Science
- ArXiv
- 2020

This work introduces a method for generating symbolic expressions to solve differential equations while leveraging deep learning training methods, and proposes a novel neural architecture for learning mathematical expressions to optimize a customizable objective. Expand

Symbolic Brittleness in Sequence Models: on Systematic Generalization in Symbolic Mathematics

- Computer Science
- ArXiv
- 2021

This work considers the problem of symbolic mathematical integration, as it requires generalizing systematically beyond the test set, and develops a methodology for evaluating generalization that takes advantage of the problem domain’s structure and access to a verifier. Expand

Pretrained Language Models are Symbolic Mathematics Solvers too!

- Computer Science, Mathematics
- ArXiv
- 2021

A sample efficient way of solving the symbolic tasks by first pretraining the transformer model with language translation and then fine-tuning the pretrained transformer model to solve the downstream task of symbolic mathematics. Expand

DO TRANSFORMERS UNDERSTAND POLYNOMIAL SIMPLIFICATION?

- 2020

Recently researchers have demonstrated that Transformers can be trained to learn symbolic tasks such as solving integration and differential equations in an end-toend fashion. In these setups, for an… Expand

Symbolically Solving Partial Differential Equations using Deep Learning

- Computer Science
- ArXiv
- 2020

A neural-based method for generating exact or approximate solutions to differential equations in the form of mathematical expressions that can be interpreted directly that is scalable, compact, and easily adaptable for a variety of tasks and configurations. Expand

A NEURAL METHOD FOR SYMBOLICALLY SOLVING PARTIAL DIFFERENTIAL EQUATIONS

- 2020

We describe a neural-based method for generating exact or approximate solutions to differential equations in the form of mathematical expressions. Unlike other neural methods, our system returns… Expand

Neurosymbolic AI: The 3rd Wave

- Computer Science
- ArXiv
- 2020

The insights provided by 20 years of neural-symbolic computing are shown to shed new light onto the increasingly prominent role of trust, safety, interpretability and accountability of AI. Expand

A I ] 1 0 D ec 2 02 0 Neurosymbolic AI : The 3 rd Wave

- 2020

Current advances in Artificial Intelligence (AI) and Machine Learning (ML) have achieved unprecedented impact across research communities and industry. Nevertheless, concerns about trust, safety,… Expand

The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence

- Computer Science
- ArXiv
- 2020

This work proposes a hybrid, knowledge-driven, reasoning-based approach, centered around cognitive models, that could provide the substrate for a richer, more robust AI than is currently possible. Expand

Unanswerable Questions About Images and Texts

- Computer Science, Medicine
- Frontiers in Artificial Intelligence
- 2020

The problem of unanswerable questions in VQA (visual question answering), in QA (textual question answering) and in AI generally are discussed. Expand

#### References

SHOWING 1-2 OF 2 REFERENCES

Deep Learning for Symbolic Mathematics

- Mathematics, Computer Science
- ICLR
- 2020

It is shown that neural networks can be surprisingly good at more elaborated tasks in mathematics, such as symbolic integration and solving differential equations, and a syntax for representing these mathematical problems, and methods for generating large datasets that can be used to train sequence-to-sequence models. Expand

Some Undecidable Problems Involving Elementary Functions of a Real Variable

- Mathematics, Computer Science
- J. Symb. Log.
- 1968

Let E be a set of expressions representing real, single valued, partially defined functions of one real variable. E * will be the set of functions represented by expressions in E . If A is an… Expand