Controlling Neural Networks with Rule Representations
- URL: http://arxiv.org/abs/2106.07804v1
- Date: Mon, 14 Jun 2021 23:28:56 GMT
- Title: Controlling Neural Networks with Rule Representations
- Authors: Sungyong Seo, Sercan O. Arik, Jinsung Yoon, Xiang Zhang, Kihyuk Sohn,
Tomas Pfister
- Abstract summary: We propose a novel training method to integrate rules into deep learning.
DeepCTRL incorporates a rule encoder into the model coupled with a rule-based objective.
It does not require retraining to adapt the rule strength -- at inference, the user can adjust it based on the desired operation point.
- Score: 48.165658432032636
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: We propose a novel training method to integrate rules into deep learning, in
a way their strengths are controllable at inference. Deep Neural Networks with
Controllable Rule Representations (DeepCTRL) incorporates a rule encoder into
the model coupled with a rule-based objective, enabling a shared representation
for decision making. DeepCTRL is agnostic to data type and model architecture.
It can be applied to any kind of rule defined for inputs and outputs. The key
aspect of DeepCTRL is that it does not require retraining to adapt the rule
strength -- at inference, the user can adjust it based on the desired operation
point on accuracy vs. rule verification ratio. In real-world domains where
incorporating rules is critical -- such as Physics, Retail and Healthcare -- we
show the effectiveness of DeepCTRL in teaching rules for deep learning.
DeepCTRL improves the trust and reliability of the trained models by
significantly increasing their rule verification ratio, while also providing
accuracy gains at downstream tasks. Additionally, DeepCTRL enables novel use
cases such as hypothesis testing of the rules on data samples, and unsupervised
adaptation based on shared rules between datasets.
Related papers
- Neural Symbolic Logical Rule Learner for Interpretable Learning [1.9526476410335776]
Rule-based neural networks stand out for enabling interpretable classification by learning logical rules for both prediction and interpretation.
We introduce the Normal Form Rule Learner (NFRL) algorithm, leveraging a selective discrete neural network, to learn rules in both Conjunctive Normal Form (CNF) and Disjunctive Normal Form (DNF)
Through extensive experiments on 11 datasets, NFRL demonstrates superior classification performance, quality of learned rules, efficiency and interpretability compared to 12 state-of-the-art alternatives.
arXiv Detail & Related papers (2024-08-21T18:09:12Z) - Learning Interpretable Rules for Scalable Data Representation and
Classification [11.393431987232425]
Rule-based Learner Representation (RRL) learns interpretable non-fuzzy rules for data representation and classification.
RRL can be easily adjusted to obtain a trade-off between classification accuracy and model complexity for different scenarios.
arXiv Detail & Related papers (2023-10-22T15:55:58Z) - Efficient Deep Reinforcement Learning Requires Regulating Overfitting [91.88004732618381]
We show that high temporal-difference (TD) error on the validation set of transitions is the main culprit that severely affects the performance of deep RL algorithms.
We show that a simple online model selection method that targets the validation TD error is effective across state-based DMC and Gym tasks.
arXiv Detail & Related papers (2023-04-20T17:11:05Z) - CT-DQN: Control-Tutored Deep Reinforcement Learning [4.395396671038298]
Control-Tutored Deep Q-Networks (CT-DQN) is a Deep Reinforcement Learning algorithm that leverages a control tutor to reduce learning time.
We validate our approach on three scenarios from OpenAI Gym: the inverted pendulum, lunar lander, and car racing.
arXiv Detail & Related papers (2022-12-02T17:59:43Z) - Deep Grey-Box Modeling With Adaptive Data-Driven Models Toward
Trustworthy Estimation of Theory-Driven Models [88.63781315038824]
We present a framework that enables us to analyze a regularizer's behavior empirically with a slight change in the neural net's architecture and the training objective.
arXiv Detail & Related papers (2022-10-24T10:42:26Z) - Adaptive Convolutional Dictionary Network for CT Metal Artifact
Reduction [62.691996239590125]
We propose an adaptive convolutional dictionary network (ACDNet) for metal artifact reduction.
Our ACDNet can automatically learn the prior for artifact-free CT images via training data and adaptively adjust the representation kernels for each input CT image.
Our method inherits the clear interpretability of model-based methods and maintains the powerful representation ability of learning-based methods.
arXiv Detail & Related papers (2022-05-16T06:49:36Z) - Stochastic Deep Model Reference Adaptive Control [9.594432031144715]
We present a Deep Neural Network-based Model Reference Adaptive Control.
Deep Model Reference Adaptive Control uses a Lyapunov-based method to adapt the output-layer weights of the DNN model in real-time.
A data-driven supervised learning algorithm is used to update the inner-layers parameters.
arXiv Detail & Related papers (2021-08-04T14:05:09Z) - Credit Assignment in Neural Networks through Deep Feedback Control [59.14935871979047]
Deep Feedback Control (DFC) is a new learning method that uses a feedback controller to drive a deep neural network to match a desired output target and whose control signal can be used for credit assignment.
The resulting learning rule is fully local in space and time and approximates Gauss-Newton optimization for a wide range of connectivity patterns.
To further underline its biological plausibility, we relate DFC to a multi-compartment model of cortical pyramidal neurons with a local voltage-dependent synaptic plasticity rule, consistent with recent theories of dendritic processing.
arXiv Detail & Related papers (2021-06-15T05:30:17Z) - Chance-Constrained Control with Lexicographic Deep Reinforcement
Learning [77.34726150561087]
This paper proposes a lexicographic Deep Reinforcement Learning (DeepRL)-based approach to chance-constrained Markov Decision Processes.
A lexicographic version of the well-known DeepRL algorithm DQN is also proposed and validated via simulations.
arXiv Detail & Related papers (2020-10-19T13:09:14Z)
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.