scholarly journals Human-Driven FOL Explanations of Deep Learning

2020 ◽  
Author(s):  
Gabriele Ciravegna ◽  
Francesco Giannini ◽  
Marco Gori ◽  
Marco Maggini ◽  
Stefano Melacci
Keyword(s):  
Deep Learning ◽  
Scientific Community ◽  
Deep Neural Networks ◽  
Order Logic ◽  
Unified Framework ◽  
First Order ◽  
Black Boxes ◽  
Classifier Performance ◽  
New Knowledge ◽  
Internal Architecture

Deep neural networks are usually considered black-boxes due to their complex internal architecture, that cannot straightforwardly provide human-understandable explanations on how they behave. Indeed, Deep Learning is still viewed with skepticism in those real-world domains in which incorrect predictions may produce critical effects. This is one of the reasons why in the last few years Explainable Artificial Intelligence (XAI) techniques have gained a lot of attention in the scientific community. In this paper, we focus on the case of multi-label classification, proposing a neural network that learns the relationships among the predictors associated to each class, yielding First-Order Logic (FOL)-based descriptions. Both the explanation-related network and the classification-related network are jointly learned, thus implicitly introducing a latent dependency between the development of the explanation mechanism and the development of the classifiers. Our model can integrate human-driven preferences that guide the learning-to-explain process, and it is presented in a unified framework. Different typologies of explanations are evaluated in distinct experiments, showing that the proposed approach discovers new knowledge and can improve the classifier performance.

scholarly journals Learning Relational Representations with Auto-encoding Logic Programs

2019 ◽  
Author(s):  
Sebastijan Dumancic ◽  
Tias Guns ◽  
Wannes Meert ◽  
Hendrik Blockeel
Keyword(s):  
Deep Learning ◽  
Relational Learning ◽  
Order Logic ◽  
First Order Logic ◽  
Relational Data ◽  
Logic Programs ◽  
Learning Methods ◽  
First Order ◽  
Relational Structures ◽  
Representation Language

Deep learning methods capable of handling relational data have proliferated over the past years. In contrast to traditional relational learning methods that leverage first-order logic for representing such data, these methods aim at re-representing symbolic relational data in Euclidean space. They offer better scalability, but can only approximate rich relational structures and are less flexible in terms of reasoning. This paper introduces a novel framework for relational representation learning that combines the best of both worlds. This framework, inspired by the auto-encoding principle, uses first-order logic as a data representation language, and the mapping between the the original and latent representation is done by means of logic programs instead of neural networks. We show how learning can be cast as a constraint optimisation problem for which existing solvers can be used. The use of logic as a representation language makes the proposed framework more accurate (as the representation is exact, rather than approximate), more flexible, and more interpretable than deep learning methods. We experimentally show that these latent representations are indeed beneficial in relational learning tasks.

scholarly journals TensorLog: A Probabilistic Database Implemented Using Deep-Learning Infrastructure

2020 ◽  
Vol 67 ◽  
pp. 285-325
Author(s):  
William Cohen ◽  
Fan Yang ◽  
Kathryn Rivard Mazaitis
Keyword(s):  
Deep Learning ◽  
High Performance ◽  
Order Logic ◽  
Parallel Processors ◽  
Probabilistic Logic ◽  
First Order ◽  
Differentiable Functions ◽  
Probabilistic Database ◽  
Learning Frameworks ◽  
Close Integration

We present an implementation of a probabilistic first-order logic called TensorLog, in which classes of logical queries are compiled into differentiable functions in a neural-network infrastructure such as Tensorflow or Theano. This leads to a close integration of probabilistic logical reasoning with deep-learning infrastructure: in particular, it enables high-performance deep learning frameworks to be used for tuning the parameters of a probabilistic logic. The integration with these frameworks enables use of GPU-based parallel processors for inference and learning, making TensorLog the first highly parallellizable probabilistic logic. Experimental results show that TensorLog scales to problems involving hundreds of thousands of knowledge-base triples and tens of thousands of examples.

scholarly journals Compositional Neural Logic Programming

2021 ◽  
Author(s):  
Son N. Tran
Keyword(s):  
Neural Networks ◽  
Logic Programming ◽  
Order Logic ◽  
First Order Logic ◽  
First Order ◽  
Forward Chaining ◽  
Symbolic Reasoning ◽  
New Knowledge ◽  
Neural Logic

This paper introduces Compositional Neural Logic Programming (CNLP), a framework that integrates neural networks and logic programming for symbolic and sub-symbolic reasoning. We adopt the idea of compositional neural networks to represent first-order logic predicates and rules. A voting backward-forward chaining algorithm is proposed for inference with both symbolic and sub-symbolic variables in an argument-retrieval style. The framework is highly flexible in that it can be constructed incrementally with new knowledge, and it also supports batch reasoning in certain cases. In the experiments, we demonstrate the advantages of CNLP in discriminative tasks and generative tasks.

Individual Reasoning within a Reasoning Community

2012 ◽  
pp. 61-86
Keyword(s):  
Decision Making ◽  
Order Logic ◽  
First Order Logic ◽  
Default Reasoning ◽  
Argumentation Schemes ◽  
Technological Systems ◽  
First Order ◽  
Deontic Reasoning ◽  
New Knowledge

In this chapter, the nature of the process that each participant engages in individually in order to contribute to collective reasoning is discussed. The design of technological systems that will best support reasoning in its communal context requires the specification of schemes for representing knowledge and for the inference of new knowledge. Further, it is also necessary to articulate a model for the process that individuals engage in when reasoning in groups. The assertion we make is that the process iteratively includes phases of engagement, individual reasoning, group coalescing, until decision making. Representations, including the classical syllogism, first order logic, default reasoning, deontic reasoning, and argumentation schemes, are surveyed to illustrate their strengths and limitations to represent individual reasoning.

A unified framework for automated person re-indentification

2020 ◽  
Vol 71 (7) ◽  
pp. 868-880
Author(s):  
Nguyen Hong-Quan ◽  
Nguyen Thuy-Binh ◽  
Tran Duc-Long ◽  
Le Thi-Lan
Keyword(s):  
Deep Learning ◽  
Video Analysis ◽  
Camera Network ◽  
Unified Framework ◽  
Person Detection ◽  
Practical Applications ◽  
Detection And Tracking ◽  
Analysis System ◽  
Bounding Boxes

Along with the strong development of camera networks, a video analysis system has been become more and more popular and has been applied in various practical applications. In this paper, we focus on person re-identification (person ReID) task that is a crucial step of video analysis systems. The purpose of person ReID is to associate multiple images of a given person when moving in a non-overlapping camera network. Many efforts have been made to person ReID. However, most of studies on person ReID only deal with well-alignment bounding boxes which are detected manually and considered as the perfect inputs for person ReID. In fact, when building a fully automated person ReID system the quality of the two previous steps that are person detection and tracking may have a strong effect on the person ReID performance. The contribution of this paper are two-folds. First, a unified framework for person ReID based on deep learning models is proposed. In this framework, the coupling of a deep neural network for person detection and a deep-learning-based tracking method is used. Besides, features extracted from an improved ResNet architecture are proposed for person representation to achieve a higher ReID accuracy. Second, our self-built dataset is introduced and employed for evaluation of all three steps in the fully automated person ReID framework.

First-Order Logic Reasoning Support for the Semantic Web

2009 ◽  
Vol 19 (12) ◽  
pp. 3091-3099 ◽  
Author(s):  
Gui-Hong XU ◽  
Jian ZHANG
Keyword(s):  
Semantic Web ◽  
Order Logic ◽  
First Order Logic ◽  
First Order ◽  
Logic Reasoning

Boolean-valued structures

2018 ◽  
Author(s):  
Tim Button ◽  
Sean Walsh
Keyword(s):  
Model Theory ◽  
Order Logic ◽  
First Order Logic ◽  
Inference Rules ◽  
Mathematical Concepts ◽  
Semantic Framework ◽  
First Order ◽  
Standard Models ◽  
Boolean Valued Model ◽  
Semantic Values

Chapters 6-12 are driven by questions about the ability to pin down mathematical entities and to articulate mathematical concepts. This chapter is driven by similar questions about the ability to pin down the semantic frameworks of language. It transpires that there are not just non-standard models, but non-standard ways of doing model theory itself. In more detail: whilst we normally outline a two-valued semantics which makes sentences True or False in a model, the inference rules for first-order logic are compatible with a four-valued semantics; or a semantics with countably many values; or what-have-you. The appropriate level of generality here is that of a Boolean-valued model, which we introduce. And the plurality of possible semantic values gives rise to perhaps the ‘deepest’ level of indeterminacy questions: How can humans pin down the semantic framework for their languages? We consider three different ways for inferentialists to respond to this question.

Categoricity and the sets

2018 ◽  
Author(s):  
Tim Button ◽  
Sean Walsh
Keyword(s):  
Set Theory ◽  
Second Order ◽  
Order Logic ◽  
First Order ◽  
The Status ◽  
Second Order Logic ◽  
Philosophy Of Set Theory ◽  
Order Set

In this chapter, the focus shifts from numbers to sets. Again, no first-order set theory can hope to get anywhere near categoricity, but Zermelo famously proved the quasi-categoricity of second-order set theory. As in the previous chapter, we must ask who is entitled to invoke full second-order logic. That question is as subtle as before, and raises the same problem for moderate modelists. However, the quasi-categorical nature of Zermelo's Theorem gives rise to some specific questions concerning the aims of axiomatic set theories. Given the status of Zermelo's Theorem in the philosophy of set theory, we include a stand-alone proof of this theorem. We also prove a similar quasi-categoricity for Scott-Potter set theory, a theory which axiomatises the idea of an arbitrary stage of the iterative hierarchy.

Sign in / Sign up

Export Citation Format

Share Document