NeuralThink: Learning Algorithms For Consistent and Efficient Extrapolation Across General Tasks
- URL: http://arxiv.org/abs/2402.15393v2
- Date: Fri, 7 Jun 2024 17:10:09 GMT
- Title: NeuralThink: Learning Algorithms For Consistent and Efficient Extrapolation Across General Tasks
- Authors: Bernardo Esteves, Miguel Vasco, Francisco S. Melo,
- Abstract summary: We propose a novel deep thinking architecture that can extrapolate, i.e., learn algorithms from smaller problems.
We show that NeuralThink consistently outperforms the prior state-of-the-art deep thinking approaches in extrapolating to larger problems.
- Score: 3.9052860539161918
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: We propose NeuralThink, a novel deep thinking architecture that can efficiently and consistently extrapolate, i.e., learn algorithms from smaller problems (in terms of observation size) and execute those algorithms in large problems. Contrary to previous deep thinking architectures, NeuralThink can be naturally applied in both same-size problems, where the input and output sizes are the same, and in different-size problems, where the size of the input and output differ. To allow for this versatility, we design NeuralThink with three main components: a recurrent module, that iteratively processes input information at different scales, a processing module, responsible for aggregating the previously processed information, and a curriculum-based training scheme, that improves the extrapolation performance of the method. To evaluate our method we introduce a set of novel different-size tasks and we show that NeuralThink consistently outperforms the prior state-of-the-art deep thinking approaches in extrapolating to larger problems, considering smaller training problems and requiring less parameters than other approaches.
Related papers
- Too Big, so Fail? -- Enabling Neural Construction Methods to Solve
Large-Scale Routing Problems [10.832715681422842]
We show that even state-of-the-art neural construction methods are outperformed by simple iterations.
We propose to use the ruin recreate principle that alternates between completely destroying a localized part of the solution and then recreating an improved variant.
arXiv Detail & Related papers (2023-09-29T09:36:37Z) - The Clock and the Pizza: Two Stories in Mechanistic Explanation of
Neural Networks [59.26515696183751]
We show that algorithm discovery in neural networks is sometimes more complex.
We show that even simple learning problems can admit a surprising diversity of solutions.
arXiv Detail & Related papers (2023-06-30T17:59:13Z) - Neural Algorithmic Reasoning Without Intermediate Supervision [21.852775399735005]
We focus on learning neural algorithmic reasoning only from the input-output pairs without appealing to the intermediate supervision.
We build a self-supervised objective that can regularise intermediate computations of the model without access to the algorithm trajectory.
We demonstrate that our approach is competitive to its trajectory-supervised counterpart on tasks from the CLRSic Algorithmic Reasoning Benchmark.
arXiv Detail & Related papers (2023-06-23T09:57:44Z) - Neural Combinatorial Optimization: a New Player in the Field [69.23334811890919]
This paper presents a critical analysis on the incorporation of algorithms based on neural networks into the classical optimization framework.
A comprehensive study is carried out to analyse the fundamental aspects of such algorithms, including performance, transferability, computational cost and to larger-sized instances.
arXiv Detail & Related papers (2022-05-03T07:54:56Z) - Adaptive Discretization in Online Reinforcement Learning [9.560980936110234]
Two major questions in designing discretization-based algorithms are how to create the discretization and when to refine it.
We provide a unified theoretical analysis of tree-based hierarchical partitioning methods for online reinforcement learning.
Our algorithms are easily adapted to operating constraints, and our theory provides explicit bounds across each of the three facets.
arXiv Detail & Related papers (2021-10-29T15:06:15Z) - Offline Model-Based Optimization via Normalized Maximum Likelihood
Estimation [101.22379613810881]
We consider data-driven optimization problems where one must maximize a function given only queries at a fixed set of points.
This problem setting emerges in many domains where function evaluation is a complex and expensive process.
We propose a tractable approximation that allows us to scale our method to high-capacity neural network models.
arXiv Detail & Related papers (2021-02-16T06:04:27Z) - Neural Function Modules with Sparse Arguments: A Dynamic Approach to
Integrating Information across Layers [84.57980167400513]
Neural Function Modules (NFM) aims to introduce the same structural capability into deep learning.
Most of the work in the context of feed-forward networks combining top-down and bottom-up feedback is limited to classification problems.
The key contribution of our work is to combine attention, sparsity, top-down and bottom-up feedback, in a flexible algorithm.
arXiv Detail & Related papers (2020-10-15T20:43:17Z) - Strong Generalization and Efficiency in Neural Programs [69.18742158883869]
We study the problem of learning efficient algorithms that strongly generalize in the framework of neural program induction.
By carefully designing the input / output interfaces of the neural model and through imitation, we are able to learn models that produce correct results for arbitrary input sizes.
arXiv Detail & Related papers (2020-07-07T17:03:02Z) - Learning to Stop While Learning to Predict [85.7136203122784]
Many algorithm-inspired deep models are restricted to a fixed-depth'' for all inputs.
Similar to algorithms, the optimal depth of a deep architecture may be different for different input instances.
In this paper, we tackle this varying depth problem using a steerable architecture.
We show that the learned deep model along with the stopping policy improves the performances on a diverse set of tasks.
arXiv Detail & Related papers (2020-06-09T07:22:01Z) - Structure preserving deep learning [1.2263454117570958]
deep learning has risen to the foreground as a topic of massive interest.
There are multiple challenging mathematical problems involved in applying deep learning.
A growing effort to mathematically understand the structure in existing deep learning methods.
arXiv Detail & Related papers (2020-06-05T10:59:09Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.