Minimal learning agents can infer unobserved variables in complex environments.
problem How to infer unobserved variables in complex environments using minimal learning agents.
method Concrete operational definition of abstract concepts, minimal architecture supporting abstraction, reinforcement learning.
result Minimal learning agents can infer the existence of unobserved variables.
Abstraction plays a key role in concept learning and knowledge discovery; this paper is concerned with computational abstraction. In particular, we study the nature of abstraction through a group-theoretic approach, formalizing it as symmetry-driven---as opposed to data-driven---hierarchical clustering. Thus, the resul…
Abstract framework for no-arbitrage concepts in topological vector lattices.
problem Generalization of no-arbitrage concepts in topological vector lattices.
method Imposing a structural condition on trading strategies and deriving abstract FTAP.
result NUPBR, NAA1, and NA1 may not be equivalent in general setting. This work abstracts deep neural networks into concept graphs for better interpretability in medical tasks.
problem Lack of interpretability in deep learning models, especially in medical domains.
method Developed a graphical representation of medical image processing models to understand concept-based reasoning.
result Extracted a concept-level graph that reveals the decision-making process of deep learning models.
We present a training system, which can provably defend significantly larger neural networks than previously possible, including ResNet-34 and DenseNet-100. Our approach is based on differentiable abstract interpretation and introduces two novel concepts: (i) abstract layers for fine-tuning the precision and scalabilit…
Examines parallels between human subjects and texts for causal inference.
problem Ambiguity and fallacies in causal inference using textual data.
method Two strategies: shifting from traits to perceptions and from concepts to parts.
result Highlights the importance of clarifying fundamental concepts.
Defines a generalized string concept for abstract root systems.
problem Generalizing the concept of strings to abstract root systems.
method Introduces a new definition for Φ-strings in abstract root systems. result Defines a new set of elements in Σ based on a given λ and subset Φ of simple roots. The seemingly infinite diversity of the natural world arises from a relatively small set of coherent rules, such as the laws of physics or chemistry. We conjecture that these rules give rise to regularities that can be discovered through primarily unsupervised experiences and represented as abstract concepts. If such r…
ARNe model excels in abstract visual reasoning tasks.
problem Abstract visual reasoning using attention mechanisms.
method Hybrid network architecture combining self-attention and relational reasoning.
result ARNe model surpasses WReN model by 11.28 ppt on PGM datasets.
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.
XIMP improves molecular property prediction by integrating multiple graph representations.
problem Graph neural networks struggle in data-scarce regimes and fail to surpass traditional methods.
method Cross-graph inter-message passing with multiple graph abstractions.
result XIMP outperforms state-of-the-art baselines across diverse molecular property tasks.
Deep learning models generate languages that lack abstract reasoning.
problem Lack of abstract reasoning in deep learning-generated languages.
method Analyzed emergent language from two multi-agent games with compositional measures.
result Deep learning solutions often fail to generalize to out-of-training examples.
Prob2Vec embeds problems for adaptive tutoring, achieving high similarity accuracy.
problem Retrieve problems with similar mathematical concepts for adaptive tutoring.
method Hierarchical problem embedding algorithm (Prob2Vec) combining abstraction and embedding steps.
result 96.88% accuracy on problem similarity test, significantly outperforming state-of-the-art sentence embedding methods.
This paper analyzes how diffusion models learn and generalize concepts.
problem Learning and generalizing concepts in compositional data-generating processes.
method Introduced a structured identity mapping (SIM) task to analyze neural network learning dynamics.
result SIM task captures key empirical observations on compositional generalization.
A new concept of causality for abstract phenomena.
problem Unclear definition of causality in real-life variables.
method Introduces 'phenomenological causality' based on elementary actions.
result Defines causal structure without hard-wired links.
After learning a concept, humans are also able to continually generalize their learned concepts to new domains by observing only a few labeled instances without any interference with the past learned knowledge. In contrast, learning concepts efficiently in a continual learning setting remains an open challenge for curr…
We introduce the concept of solenoid as an abstract laminated space. We do a thorough study of solenoids, leading to the notion of ergodic and uniquely ergodic solenoids. We define generalized currents associated with immersions of oriented solenoids with a transversal measure into smooth manifolds, generalizing Ruelle…
We describe a general framework for measuring risks, where the risk measure takes values in an abstract cone. It is shown that this approach naturally includes the classical risk measures and set-valued risk measures and yields a natural definition of vector-valued risk measures. Several main constructions of risk meas…
The study explores highly supersymmetric backgrounds in 11D supergravity.
problem Understanding and constructing highly supersymmetric backgrounds in 11D supergravity.
method Definition of abstract symbols and a strong version of the Reconstruction Theorem, proposing a strategy to construct backgrounds, and providing an example with detailed computation.
result Bijective correspondence between highly supersymmetric backgrounds and abstract symbols, and a classical supersymmetry gap result.
We develop a method to learn abstract causal graphs from interventional data.
problem Estimating causal models at fine granularity is impractical or undesirable.
method Novel graphical identifiability results and an efficient algorithm.
result Directly learns abstract causal graphs from interventional data.
The abstract reviews financial concepts using physics.
problem Financial pricing and risk management.
method Discrete time formalism, path integral, Green's function formulas.
result Formulas for pricing and risk mitigation methods.
Solving complex, temporally-extended tasks is a long-standing problem in reinforcement learning (RL). We hypothesize that one critical element of solving such problems is the notion of compositionality. With the ability to learn concepts and sub-skills that can be composed to solve longer tasks, i.e. hierarchical RL, w…
The abstract aims to generalize classical curve concepts to uniquely define complex curves.
problem Lack of sufficient information to distinguish between different curves.
method Generalizing classical concepts of curvature and torsion to higher algebraic curvatures.
result Each analytic branch of a complex curve is uniquely defined by higher algebraic curvatures.
Defines tensor eigenvalues and singular values without basis, simplifying analysis.
problem Defines tensor eigenvalues and singular values without basis.
method Intrinsic definition of tensor eigenvalues and singular values using concepts from pure mathematics.
result Shows the relationship between tensor analysis and pure mathematics.
Formalizes concepts as latent variables in hierarchical models for high-dimensional data.
problem Lack of formalization and theoretical insights for learning discrete concepts from high-dimensional data.
method Formalizes concepts as latent causal variables in a hierarchical model, formulates conditions for concept identification.
result Conditions for identifying latent hierarchical models in unsupervised data, handling complex structures and high-dimensional data.
This paper improves disentanglement in VAEs by progressively learning hierarchical representations.
problem Compromised disentanglement in VAEs due to high-level abstraction extraction.
method Progressive learning of independent hierarchical representations from high to low levels.
result Improved disentanglement demonstrated on two benchmark datasets using new metrics.
This work explains how linear representations in large language models arise from training objectives and gradient descent.
problem Understanding the origins of linear representations in large language models.
method A latent variable model to abstract and formalize concept dynamics, combined with analysis of the softmax cross-entropy objective and gradient descent.
result Linear representations emerge when learning from data matching the latent variable model, and this simple structure suffices to yield linear representations.
Proposes new methods for interpreting document classification models.
problem Interpretation fragility of attention-based neural networks.
method Corpus-level and concept-based explanation methods using attention weights.
result Extracts semantically meaningful keywords and concepts for model predictions.
New cohomology groups generalize Euler number for Lie superalgebras.
problem Generalizing cohomology groups for Lie superalgebras.
method Abstracted Poisson cohomology groups to Poisson-like cohomology groups for general Lie superalgebras.
result De Rham cohomology groups match Poisson-like cohomology groups for differential forms.
MXGNet tackles visual reasoning tasks using graph neural networks.
problem Abstract reasoning, especially in the visual domain, is challenging for AI.
method Combines object-level representations, graph neural networks, and multiplex graphs.
result Achieves state-of-the-art accuracy on Euler Diagram Syllogisms and outperforms state-of-the-art models on RPM datasets.
Unified model learns concepts across domains like left and right.
problem Limited generalization of language concepts in inference-only models.
method Logic-Enhanced Foundation Model (LEFT) with a differentiable, domain-independent program executor.
result LEFT flexibly learns and reasons with concepts across 2D images, 3D scenes, human motions, and robotic manipulation.
The paper discusses iCurrency concept and Libra's potential as a contender.
problem Analyzing Libra's potential as a contender for iCurrency.
method Analyzed Libra proposal, discussed monetary policy issues.
result Libra faces challenges in maintaining exchange rate stability.
Deep generative models are reported to be useful in broad applications including image generation. Repeated inference between data space and latent space in these models can denoise cluttered images and improve the quality of inferred results. However, previous studies only qualitatively evaluated image outputs in data…
In this semi-tutorial paper, we first review the information-theoretic approach to account for the computational costs incurred during the search for optimal actions in a sequential decision-making problem. The traditional (MDP) framework ignores computational limitations while searching for optimal policies, essential…
The article provides a modest survey of the absolute theory of general systems of (partial) differential equations. The equations are relieved of all additional structures and subject to quite arbitrary change of the variables. An abstract mathematical theory in the Bourbaki sense with its own concepts and technical to…
Improved scalability and interpretability in training data attribution.
problem Identifying which training data drives specific behaviors, especially unintended ones.
method Leveraging interpretable structures within the model to attribute model behavior to semantic directions, not individual test examples.
result Simple probe-based attribution methods are first-order approximations of Concept Influence that achieve comparable performance while being over an order-of-magnitude faster.
Abstract: Surveying connections between ML and Control Theory.
problem Addressing the intersection of Machine Learning and Control Theory.
method Develops connections through reinforcement learning, supervised learning, deep learning, and stochastic gradient descent.
result Machine Learning and Control Theory are interconnected, with ML solving large control problems and Control Theory providing tools for ML.
Text documents can be described by a number of abstract concepts such as semantic category, writing style, or sentiment. Machine learning (ML) models have been trained to automatically map documents to these abstract concepts, allowing to annotate very large text collections, more than could be processed by a human in …
Machine learning is usually defined in behaviourist terms, where external validation is the primary mechanism of learning. In this paper, I argue for a more holistic interpretation in which finding more probable, efficient and abstract representations is as central to learning as performance. In other words, machine le…
Survey on discrete curvature concepts for polygons and polyhedral surfaces.
problem Defining curvature for discrete structures like polygons and polyhedral surfaces.
method Explains curvature notions for polygons, polyhedral surfaces, and abstract polyhedral manifolds.
result Discrete curvature theorems parallel classical theorems in differential geometry.
A number of recent approaches to policy learning in 2D game domains have been successful going directly from raw input images to actions. However when employed in complex 3D environments, they typically suffer from challenges related to partial observability, combinatorial exploration spaces, path planning, and a scarc…
DORA analyzes deep neural networks' internal representations to detect spurious correlations.
problem Detecting spurious correlations in deep neural networks' internal representations.
method DORA uses Extreme-Activation (EA) distance measure to assess representation similarities.
result Identifies internal representations capable of detecting spurious correlations.
CGAs estimate team performance from data, simplifying SV computation.
problem Predicting and rewarding team performance using game theory.
method Cooperative game abstractions (CGAs) for estimating characteristic functions from data.
result CGAs enable linear-time computation of Shapley Value for team contributions.
Machine learning has made major advances in categorizing objects in images, yet the best algorithms miss important aspects of how people learn and think about categories. People can learn richer concepts from fewer examples, including causal models that explain how members of a category are formed. Here, we explore the…
Deep CNNs are known to exhibit the following peculiarity: on the one hand they generalize extremely well to a test set, while on the other hand they are extremely sensitive to so-called adversarial perturbations. The extreme sensitivity of high performance CNNs to adversarial examples casts serious doubt that these net…
It is easy for people to imagine what a man with pink hair looks like, even if they have never seen such a person before. We call the ability to create images of novel semantic concepts visually grounded imagination. In this paper, we show how we can modify variational auto-encoders to perform this task. Our method use…
Abstracts index for ML4H workshop at NeurIPS 2019.
problem No specific problem stated; index of accepted abstracts.
method Not specified; index of accepted abstracts.
result No specific result stated.
Abstraction is a fundamental part when learning behavioral models of systems. Usually the process of abstraction is manually defined by domain experts. This paper presents a method to perform automatic abstraction for network protocols. In particular a weakly supervised clustering algorithm is used to build an abstract…