Develops logic programs for explaining classification model decisions.
problem Creating explanations for decisions made by classification models.
method Answer-set programs for computing counterfactual interventions.
result Maximum responsibility causal explanations can be computed.
The paper explains deep neural network predictions using logical proxies.
problem Creating understandable explanations for deep neural network predictions.
method Randomized propositionalization and Bayes-like approach to identify logical proxies.
result Models in first-order logic can approximate DRM's predictions in local regions.
Enhances visual explanations with logical rules for complex concepts.
problem Lack of explanatory power for deep learning models, especially for relational concepts.
method Combining LIME for highlighting and Aleph for logic rules.
result Generated relational rules can be explicitly linked to input images.
The paper shows averaging gradients leads to memorization, proposing an alternative algorithm to focus on invariances.
problem The principle that 'good explanations are hard to vary' in deep learning is investigated.
method Formalizing consistency for loss surface minima, proposing an alternative algorithm based on logical AND.
result The alternative algorithm prevents memorization and focuses on invariances.
Bayesian interpolants explain neural network inferences concisely.
problem Understanding neural network inferences.
method Adapting Craig interpolants for neural networks.
result Produces precise, understandable explanations.
In recent years, a number of artificial intelligent services have been developed such as defect detection system or diagnosis system for customer services. Unfortunately, the core in these services is a black-box in which human cannot understand the underlying decision making logic, even though the inspection of the lo…
RDLI integrates domain logic and context grounding to detect crypto anomalies under scarce labels.
problem Extreme label scarcity and evasion strategies in crypto networks.
method Relational Domain Logic Integration (RDLI) with Retrieval Grounded Context (RGC).
result RDLI outperforms GNN baselines by 28.9% in F1 score under 0.01% label scarcity.
Describes explaining neurons in deep representations using compositional logical concepts.
problem Interpreting neuron behavior in deep neural networks.
method Identifying compositional logical concepts that closely approximate neuron behavior.
result Compositional explanations provide insights into model performance and allow for adversarial example creation.
OutlierTree detects outliers using decision trees and provides explanations.
problem Detecting outliers in data while providing understandable explanations.
method Supervised decision tree splits with 1-d confidence intervals.
result Human-readable explanations for outlier detection.
Improves deep learning interpretability through logical network construction.
problem Difficulty in understanding and explaining deep neural networks, especially convolution neural networks.
method Proposes a method to construct a logical network based on fully connected layers to improve interpretability.
result Improves the interpretability of deep learning processes and models.
A-NeSI scales approximate inference for probabilistic neurosymbolic learning.
problem Combining neural networks with symbolic reasoning for scalable inference.
method A-NeSI: a new framework for PNL using neural networks for approximate inference.
result A-NeSI achieves scalable approximate inference without semantic changes.
New method compares feature importance and rule extraction for text data interpretability.
problem Unexpected differences in explanations from feature importance and rule extraction methods.
method Proposes a new approach to compare explanations from different methods.
result Different methods can lead to unexpected explanations, even for simple models.
Proposes approximating computationally expensive explainability techniques using conformal regression.
problem Computational expense of score-based explainability techniques limits their applicability in time-critical contexts.
method Uses conformal prediction framework to approximate SHAP and TreeSHAP explanations.
result Significantly improves execution time and produces tight validity guarantees.
The paper explores how different network architectures learn logical functions under GOTU, finding that a min-degree-interpolator is learned.
problem Learning logical functions with a focus on generalization on the unseen.
method Study of different network architectures trained by SGD under GOTU.
result For sparse functions and certain network models, a min-degree-interpolator is learned on the unseen.
Fairwashing occurs when machine learning models are made to appear fair through rationalization.
problem Rationalizing unfair black-box models to appear fair.
method LaundryML uses a regularized rule list enumeration algorithm to find fair rule lists approximating an unfair model.
result It is possible to systematically rationalize decisions from unfair black-box models using model and outcome explanations.
New algorithm generates counterfactual explanations for diverse models without restrictions.
problem Supporting consequential decisions with understandable explanations for predictive models.
method Solves satisfiability problems using logic formulae for model-agnostic, diverse explanations.
result Generates diverse, plausible counterfactuals at provably optimal distances.
Manifold visualizes machine learning model outcomes without accessing their internal logic.
problem Lack of generic frameworks for interpreting and diagnosing different machine learning models.
method Generic framework that observes inputs and outputs, not model's internal logic.
result Supports interpretation, debugging, and comparison of machine learning models in a transparent and interactive manner.
New saliency evaluations focus on completeness and soundness, improving explanations.
problem Current saliency evaluations focus on completeness but ignore soundness.
method Introduces new intrinsic evaluation metrics based on completeness and soundness.
result Simple saliency method matches or outperforms prior methods in new evaluations.
AlphaLogics mines market logic to generate interpretable alpha factors.
problem Complex, opaque alpha factors from factor mining overlook market logic.
method Market Logic Mining, Factor Generation and Optimization, Market Logic Generation and Optimization.
result AlphaLogics improves predictive metrics and risk-adjusted returns over baselines.
NLN learns logical reasoning from neural networks.
problem Lack of logical reasoning in deep neural networks.
method Dynamic neural architecture that builds computational graph based on logical expressions, learns logical operations as neural modules, conducts propositional logical reasoning.
result NLN significantly outperforms state-of-the-art models on collaborative filtering and personalized recommendation tasks.
Paper studies a universal logical operator for deep networks, improving interpretability.
problem Learning a universal logical operator for deep convolution networks without manual prescription.
method Exploration of different logical operators (AND, OR, XOR) and learning a universal one.
result Insightful observations lead to a novel logical interpretation of deep convolution networks.
A new method for variable importance measures without impossible data.
problem Using impossible data for variable importance measures in black box models.
method Cohort Shapley, a method grounded in economic game theory using only observed data.
result Cohort Shapley provides a more trustworthy explanation of black box models' decisions.
NLM combines neural networks and logic programming for complex reasoning.
problem Complex reasoning tasks involving logic and properties.
method Neural-symbolic architecture combining neural networks and logic programming.
result NLM achieves perfect generalization on various tasks.
Polyhedral semantics for intermediate logics; Nerve Criterion ensures completeness.
problem Characterize polyhedrally-complete intermediate logics.
method Developed Nerve Criterion to characterize polyhedrally-complete logics combinatorially.
result Nerve Criterion provides a necessary and sufficient condition for polyhedrally-completeness.
Logical scaffolds enhance AI software quality.
problem Improving AI component quality in software.
method Logical scaffolds as a method to improve AI components.
result Logical scaffolds can improve AI beyond perception systems.
Boolean logic used for neural network training and inference, with convergence analysis.
problem Discrete optimization in neural networks with Boolean logic.
method Boolean logic backpropagation with convergence analysis.
result First convergence analysis for Boolean logic in neural networks.
Graph neural networks improve logic reasoning for large datasets.
problem Combining logic reasoning and probabilistic inference for large datasets.
method Exploring Graph Neural Networks (GNNs) for Markov Logic Networks (MLN) to improve probabilistic logic inference.
result ExpressGNN, a more expressive variant of GNN, can perform effective probabilistic logic inference and scale to large datasets.
Develops deep learning for logical code segmentation.
problem Lack of logically segmented source code.
method Novel deep learning approach to generate logical code segments.
result Improves software analysis tasks like commenting, bug detection, and code synthesis.
pLogicNet combines logic rules and embeddings for efficient knowledge graph reasoning.
problem Efficiently predicting missing facts in knowledge graphs.
method Combines Markov Logic Networks with knowledge graph embeddings using variational EM algorithm.
result pLogicNet outperforms traditional methods on multiple knowledge graphs.
Graph neural networks struggle with proving unsatisfiability in complex logical formulas.
problem Proving unsatisfiability in complex logical formulas.
method Investigating the limitations of graph neural networks in logical reasoning tasks.
result Graph neural networks may fail in certifying unsatisfiability in Boolean formulae.
Neural Logic Reasoning integrates deep learning and symbolic logic for better prediction tasks.
problem Lack of cognitive reasoning in deep neural networks limits their ability to solve complex prediction tasks.
method Proposes Logic-Integrated Neural Network (LINN) that learns logical operations and conducts propositional logical reasoning.
result LINN significantly outperforms state-of-the-art recommendation models in Top-K recommendation.
Combines neural networks and logic circuits for interpretable, accurate, and cost-effective learning.
problem Lack of generalizability and interpretability in neural networks and high hardware cost in logic circuits.
method Trains a neural network, then translates it to random forests, and finally to AND-Inverter logic.
result The pipeline maintains greater accuracy and minimizes logic complexity.
FPGA-based logic architecture speeds up GBDT training 259x.
problem Training efficiency and power consumption in GBDT models.
method Implemented logic architecture on FPGA, compared with software libraries.
result Training speed 26-259x faster, power efficiency 90-1,104x higher.
Integrates deep learning and logic reasoning for intelligent agents.
problem Combining deep learning and logic reasoning for robust decision-making.
method Deep Logic Models integrating deep learners and logic reasoning.
result Proposed models outperform other approaches in joint learning and inference.
The purpose of this paper is to discuss how topology and geometry provide, in many instances, the connective tissue that enables logical comprehension. We illustrate this theme with many examples including Venn diagrams, knot diagrams, knot-logical diagrams and an arrow of reference that elucidates self-reference and G…
Kernel for STL formulae enables machine learning in temporal logic.
problem Lack of a kernel for STL formulae.
method Define a kernel for STL formulae and embed them into a Hilbert space.
result Kernel-based machine learning algorithms can now be applied to STL formulae.
Transformers learn to predict temporal logic solutions from classical solver outputs.
problem Training neural networks on logic problem solutions for verification.
method Training a Transformer on generated training data from classical solvers, focusing on one solution per formula.
result Transformers can predict correct solutions to temporal logic problems, even to unseen benchmarks.
New theory of co~events resolves debates between Bayesianists and frequentists.
problem Fierce debates between Bayesianists and frequentists over Bayesian scheme.
method Developed new co~event axiomatics and theory to study experience and chance as a single co~event.
result Demonstrated effectiveness of new theory in resolving debates over Bayesian scheme.
Study logical generalization in GNNs using a new benchmark.
problem Understanding how GNNs adapt to new logical tasks.
method Developed GraphLog benchmark suite for logical tasks, evaluated GNNs in supervised, pretraining, and continual learning settings.
result Logical diversity during training affects GNNs' ability to generalize.
A caching mechanism improves sequence to logical form generation accuracy.
problem Generating logical forms from natural language sequences.
method Proposes a caching mechanism to increase output probability of source input tokens and weigh them based on context.
result Improves sequence/token-level accuracy on sequence to logical form tasks.
2-simplicial Transformer enhances logical reasoning in reinforcement learning.
problem Logical reasoning in reinforcement learning.
method Introduces 2-simplicial Transformer with higher-dimensional attention and tensor product updates. result Shows effectiveness of 2-simplicial Transformer for logical reasoning. Proposes NLRL for enhancing neural networks' interpretability.
problem Deep neural networks lack interpretability for humans.
method Introduces neural logic rule layers (NLRL) to represent arbitrary logic rules.
result NLRL-enhanced neural networks can learn complex logic and arithmetic.
Paper introduces a new framework combining deep learning and logic for relational data.
problem Scalability and flexibility of deep learning methods for relational data.
method Combines auto-encoding principle with first-order logic and logic programs.
result Latent representations are more accurate, flexible, and interpretable.
NMLNs use neural networks to learn relational structure from data.
problem Learning implicit rules from data without explicit logic rules.
method Combines Markov logic with neural networks to learn relational structure.
result NMLNs can predict in settings without explicit logic rules.
Responds to comments on Bayesian Logic Regression algorithm, provides extensions and tutorial.
problem Improving Bayesian Logic Regression algorithm and its applications.
method Summarizes and responds to discussants' comments, provides extensions and tutorial.
result Extensions and tutorial for the Bayesian Logic Regression algorithm.
Tensor logic aims to unify AI types with scalable and transparent features.
problem Lack of a unified AI programming language with scalability and transparency.
method Introduces tensor logic, a new AI language based on tensor equations.
result Tensor logic enables key AI forms like transformers, formal reasoning, and graphical models.
Logical neural networks solve mazes by filling dead ends, but not all methods generalize well.
problem Understanding how logical neural networks extrapolate solutions to mazes.
method Examined recurrent and implicit neural networks trained on maze-solving tasks.
result Models fail to generalize well to diverse maze sizes, suggesting limitations in learning scalable algorithms.
In 1938, Tarski proved that a formula is not intuitionistically valid if, and only if, it has a counter-model in the Heyting algebra of open sets of some topological space. In fact, Tarski showed that any Euclidean space R^n with n >= 1 suffices, as does e.g. the Cantor space. In particular, intuitionistic logic cannot…