Search Results for author:
Found 47 papers, 18 papers with code
Spurious Correlations in Concept Drift: Can Explanatory Interaction Help?
Long-running machine learning models face the issue of concept drift (CD), whereby the data distribution changes over time, compromising prediction performance.
Perks and Pitfalls of Faithfulness in Regular, Self-Explainable and Domain Invariant GNNs
As Graph Neural Networks (GNNs) become more pervasive, it becomes paramount to build robust tools for computing explanations of their predictions.
A Benchmark Suite for Systematically Evaluating Reasoning Shortcuts
The advent of powerful neural classifiers has increased interest in problems that require both learning and reasoning.
Semantic Loss Functions for Neuro-Symbolic Structured Prediction
We discuss the semantic loss, which injects knowledge about such structure, defined symbolically, into training by minimizing the network's violation of such dependencies, steering the network towards predicting distributions satisfying the underlying structure.
Towards Logically Consistent Language Models via Probabilistic Reasoning
Large language models (LLMs) are a promising venue for natural language understanding and generation tasks.
Learning To Guide Human Decision Makers With Vision-Language Models
As a remedy, we introduce learning to guide (LTG), an alternative framework in which - rather than taking control from the human expert - the machine provides guidance useful for decision making, and the human is entirely responsible for coming up with a decision.
BEARS Make Neuro-Symbolic Models Aware of their Reasoning Shortcuts
Neuro-Symbolic (NeSy) predictors that conform to symbolic knowledge - encoding, e. g., safety constraints - can be affected by Reasoning Shortcuts (RSs): They learn concepts consistent with the symbolic knowledge by exploiting unintended semantics.
Interpretability is in the Mind of the Beholder: A Causal Framework for Human-interpretable Representation Learning
This allows us to derive a principled notion of alignment between the machine representation and the vocabulary of concepts understood by the human.
How Faithful are Self-Explainable GNNs?
Self-explainable deep neural networks are a recent class of models that can output ante-hoc local explanations that are faithful to the model's reasoning, and as such represent a step forward toward filling the gap between expressiveness and interpretability.
Learning to Guide Human Experts via Personalized Large Language Models
In learning to defer, a predictor identifies risky decisions and defers them to a human expert.
Not All Neuro-Symbolic Concepts Are Created Equal: Analysis and Mitigation of Reasoning Shortcuts
Neuro-Symbolic (NeSy) predictive models hold the promise of improved compliance with given constraints, systematic generalization, and interpretability, as they allow to infer labels that are consistent with some prior knowledge by reasoning over high-level concepts extracted from sub-symbolic inputs.
Neuro-Symbolic Reasoning Shortcuts: Mitigation Strategies and their Limitations
This setup offers clear advantages in terms of consistency to symbolic prior knowledge, and is often believed to provide interpretability benefits in that - by virtue of complying with the knowledge - the learned concepts can be better understood by human stakeholders.
Neuro-Symbolic Continual Learning: Knowledge, Reasoning Shortcuts and Concept Rehearsal
We introduce Neuro-Symbolic Continual Learning, where a model has to solve a sequence of neuro-symbolic tasks, that is, it has to map sub-symbolic inputs to high-level concepts and compute predictions by reasoning consistently with prior knowledge.
Leveraging Explanations in Interactive Machine Learning: An Overview
Explanations have gained an increasing level of interest in the AI and Machine Learning (ML) communities in order to improve model transparency and allow users to form a mental model of a trained ML model.
Semantic Probabilistic Layers for Neuro-Symbolic Learning
We design a predictive layer for structured-output prediction (SOP) that can be plugged into any neural network guaranteeing its predictions are consistent with a set of predefined symbolic constraints.
GlanceNets: Interpretabile, Leak-proof Concept-based Models
There is growing interest in concept-based models (CBMs) that combine high-performance and interpretability by acquiring and reasoning with a vocabulary of high-level concepts.
Concept-level Debugging of Part-Prototype Networks
We propose ProtoPDebug, an effective concept-level debugger for ProtoPNets in which a human supervisor, guided by the model's explanations, supplies feedback in the form of what part-prototypes must be forgotten or kept, and the model is fine-tuned to align with this supervision.
Personalized Algorithmic Recourse with Preference Elicitation
It is paramount that the sequence of actions does not require too much effort for users to implement.
Machine Learning for Combinatorial Optimisation of Partially-Specified Problems: Regret Minimisation as a Unifying Lens
The challenge is to learn them from available data, while taking into account a set of hard constraints that a solution must satisfy, and that solving the optimisation problem (esp.
Lifelong Personal Context Recognition
We focus on the development of AIs which live in lifelong symbiosis with a human.
Efficient and Reliable Probabilistic Interactive Learning with Structured Outputs
In this position paper, we study interactive learning for structured output spaces, with a focus on active learning, in which labels are unknown and must be acquired, and on skeptical learning, in which the labels are noisy and may need relabeling.
Learning MAX-SAT from Contextual Examples for Combinatorial Optimisation
Combinatorial optimisation problems are ubiquitous in artificial intelligence.
Machine Learning for Utility Prediction in Argument-Based Computational Persuasion
and (2) How can we identify for a new user the best utility function from amongst those that we have learned?
Toward a Unified Framework for Debugging Concept-based Models
In this paper, we tackle interactive debugging of "gray-box" concept-based models (CBMs).
Learning Mixed-Integer Linear Programs from Contextual Examples
In this paper, we study the problem of acquiring MILPs from contextual examples, a novel and realistic setting in which examples capture solutions and non-solutions within a specific context.
Interactive Label Cleaning with Example-based Explanations
We tackle sequential learning under label noise in applications where a human supervisor can be queried to relabel suspicious examples.
A Compositional Atlas of Tractable Circuit Operations for Probabilistic Inference
Circuit representations are becoming the lingua franca to express and reason about tractable generative and discriminative models.
Neuro-Symbolic Constraint Programming for Structured Prediction
We propose Nester, a method for injecting neural networks into constrained structured predictors.
Human-in-the-loop Handling of Knowledge Drift
Motivated by this, we introduce TRCKD, a novel approach that combines automated drift detection and adaptation with an interactive stage in which the user is asked to disambiguate between different kinds of KD.
A Compositional Atlas of Tractable Circuit Operations: From Simple Transformations to Complex Information-Theoretic Queries
Circuit representations are becoming the lingua franca to express and reason about tractable generative and discriminative models.
Bandits for Learning to Explain from Explanations
We introduce Explearn, an online algorithm that learns to jointly output predictions and explanations for those predictions.
Multi-Modal Subjective Context Modelling and Recognition
Second, existing models typically assume that context is objective, whereas in most applications context is best viewed from the user's perspective.
Learning in the Wild with Incremental Skeptical Gaussian Processes
The ability to learn from human supervision is fundamental for personal assistants and other interactive applications of AI.
Machine Guides, Human Supervises: Interactive Learning with Global Explanations
Compared to other explanatory interactive learning strategies, which are machine-initiated and rely on local explanations, XGL is designed to be robust against cases in which the explanations supplied by the machine oversell the classifier's quality.
Efficient Generation of Structured Objects with Constrained Adversarial Networks
Generative Adversarial Networks (GANs) struggle to generate structured objects like molecules and game maps.
Toward Machine-Guided, Human-Initiated Explanatory Interactive Learning
This biases the "narrative" presented by the machine to the user. We address this narrative bias by introducing explanatory guided learning, a novel interactive learning strategy in which: i) the supervisor is in charge of choosing the query instances, while ii) the machine uses global explanations to illustrate its overall behavior and to guide the supervisor toward choosing challenging, informative instances.
Human-Machine Collaboration for Democratizing Data Science
Everybody wants to analyse their data, but only few posses the data science expertise to to this.
Making deep neural networks right for the right scientific reasons by interacting with their explanations
Deep neural networks have shown excellent performances in many real-world applications.
Does Symbolic Knowledge Prevent Adversarial Fooling?
Arguments in favor of injecting symbolic knowledge into neural architectures abound.
Automating Personnel Rostering by Learning Constraints Using Tensors
Many problems in operations research require that constraints be specified in the model.
"Why Should I Trust Interactive Learners?" Explaining Interactive Queries of Classifiers to Users
Although interactive learning puts the user into the loop, the learner remains mostly a black box for the user.
Decomposition Strategies for Constructive Preference Elicitation
We propose a decomposition technique for large preference-based decision problems relying exclusively on inference and feedback over partial configurations.
Constructive Preference Elicitation over Hybrid Combinatorial Spaces
The preferences are typically learned by querying the user for choice feedback over pairs or sets of objects.
Coactive Critiquing: Elicitation of Preferences and Features
When faced with complex choices, users refine their own preference criteria as they explore the catalogue of options.
Constructive Preference Elicitation by Setwise Max-margin Learning
In this paper we propose an approach to preference elicitation that is suitable to large configuration spaces beyond the reach of existing state-of-the-art approaches.
Structured Learning Modulo Theories
The main idea is to leverage a state-of-the-art generalized Satisfiability Modulo Theory solver for implementing the inference and separation oracles of Structured Output SVMs.
Hybrid SRL with Optimization Modulo Theories
Generally speaking, the goal of constructive learning could be seen as, given an example set of structured objects, to generate novel objects with similar properties.
