Abstraction Alignment: Comparing Model-Learned and Human-Encoded Conceptual Relationships

• URL: http://arxiv.org/abs/2407.12543v2
• Date: Thu, 13 Feb 2025 20:47:20 GMT
• Title: Abstraction Alignment: Comparing Model-Learned and Human-Encoded Conceptual Relationships
• Authors: Angie Boggust, Hyemin Bang, Hendrik Strobelt, Arvind Satyanarayan,
• Abstract summary: We introduce abstraction alignment, a methodology to compare model behavior against formal human knowledge. Abstraction alignment externalizes domain-specific human knowledge as an abstraction graph. Users can test alignment hypotheses, such as which human concepts the model has learned.
• Score: 26.503178592074757
• License:
• Abstract: While interpretability methods identify a model's learned concepts, they overlook the relationships between concepts that make up its abstractions and inform its ability to generalize to new data. To assess whether models' have learned human-aligned abstractions, we introduce abstraction alignment, a methodology to compare model behavior against formal human knowledge. Abstraction alignment externalizes domain-specific human knowledge as an abstraction graph, a set of pertinent concepts spanning levels of abstraction. Using the abstraction graph as a ground truth, abstraction alignment measures the alignment of a model's behavior by determining how much of its uncertainty is accounted for by the human abstractions. By aggregating abstraction alignment across entire datasets, users can test alignment hypotheses, such as which human concepts the model has learned and where misalignments recur. In evaluations with experts, abstraction alignment differentiates seemingly similar errors, improves the verbosity of existing model-quality metrics, and uncovers improvements to current human abstractions.

Related papers

• Exploring the Trade-off Between Model Performance and Explanation Plausibility of Text Classifiers Using Human Rationales [3.242050660144211]
  Saliency post-hoc explainability methods are important tools for understanding increasingly complex NLP models. We present a methodology for incorporating rationales, which are text annotations explaining human decisions, into text classification models.
  arXiv  Detail & Related papers  (2024-04-03T22:39:33Z)
• How to Handle Sketch-Abstraction in Sketch-Based Image Retrieval? [120.49126407479717]
  We propose a sketch-based image retrieval framework capable of handling sketch abstraction at varied levels. For granularity-level abstraction understanding, we dictate that the retrieval model should not treat all abstraction-levels equally. Our Acc.@q loss uniquely allows a sketch to narrow/broaden its focus in terms of how stringent the evaluation should be.
  arXiv  Detail & Related papers  (2024-03-11T23:08:29Z)
• Neural Causal Abstractions [63.21695740637627]
  We develop a new family of causal abstractions by clustering variables and their domains. We show that such abstractions are learnable in practical settings through Neural Causal Models. Our experiments support the theory and illustrate how to scale causal inferences to high-dimensional settings involving image data.
  arXiv  Detail & Related papers  (2024-01-05T02:00:27Z)
• The Relational Bottleneck as an Inductive Bias for Efficient Abstraction [3.19883356005403]
  We show that neural networks are constrained via their architecture to focus on relations between perceptual inputs, rather than the attributes of individual inputs. We review a family of models that employ this approach to induce abstractions in a data-efficient manner.
  arXiv  Detail & Related papers  (2023-09-12T22:44:14Z)
• Quantifying Consistency and Information Loss for Causal Abstraction Learning [16.17846886492361]
  We introduce a family of interventional measures that an agent may use to evaluate such a trade-off. We consider four measures suited for different tasks, analyze their properties, and propose algorithms to evaluate and learn causal abstractions.
  arXiv  Detail & Related papers  (2023-05-07T19:10:28Z)
• Does Deep Learning Learn to Abstract? A Systematic Probing Framework [69.2366890742283]
  Abstraction is a desirable capability for deep learning models, which means to induce abstract concepts from concrete instances and flexibly apply them beyond the learning context. We introduce a systematic probing framework to explore the abstraction capability of deep learning models from a transferability perspective.
  arXiv  Detail & Related papers  (2023-02-23T12:50:02Z)
• Abstract Interpretation for Generalized Heuristic Search in Model-Based Planning [50.96320003643406]
  Domain-general model-based planners often derive their generality by constructing searchs through the relaxation of symbolic world models. We illustrate how abstract interpretation can serve as a unifying framework for these abstractions, extending the reach of search to richer world models. Theses can also be integrated with learning, allowing agents to jumpstart planning in novel world models via abstraction-derived information.
  arXiv  Detail & Related papers  (2022-08-05T00:22:11Z)
• Towards Computing an Optimal Abstraction for Structural Causal Models [16.17846886492361]
  We focus on the problem of learning abstractions. We suggest a concrete measure of information loss, and we illustrate its contribution to learning new abstractions.
  arXiv  Detail & Related papers  (2022-08-01T14:35:57Z)
• Translational Concept Embedding for Generalized Compositional Zero-shot Learning [73.60639796305415]
  Generalized compositional zero-shot learning means to learn composed concepts of attribute-object pairs in a zero-shot fashion. This paper introduces a new approach, termed translational concept embedding, to solve these two difficulties in a unified framework.
  arXiv  Detail & Related papers  (2021-12-20T21:27:51Z)
• Towards a Mathematical Theory of Abstraction [0.0]
  We provide a precise characterisation of what an abstraction is and, perhaps more importantly, suggest how abstractions can be learnt directly from data. Our results have deep implications for statistical inference and machine learning and could be used to develop explicit methods for learning precise kinds of abstractions directly from data.
  arXiv  Detail & Related papers  (2021-06-03T13:23:49Z)
• A Diagnostic Study of Explainability Techniques for Text Classification [52.879658637466605]
  We develop a list of diagnostic properties for evaluating existing explainability techniques. We compare the saliency scores assigned by the explainability techniques with human annotations of salient input regions to find relations between a model's performance and the agreement of its rationales with human ones.
  arXiv  Detail & Related papers  (2020-09-25T12:01:53Z)

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.