Characterizes concept classes for optimistic online learning.
problem Understanding minimal assumptions for online learnability.
method Investigates two questions about concept classes' learnability.
result Characterizes all concept classes for optimistically universal online learnability.
Study expands multiclass classification models with new rates and partial concept classes.
problem Multiclass classification with a bounded number of labels under various conditions.
method Extends traditional PAC model to distribution-dependent and data-dependent learning rates, characterizes optimal rates for universal and partial concept classes.
result Characterizes three types of learning rates (exponential, linear, arbitrarily slow) for fixed distributions and complexity measures for partial concept classes.
ECBMs unify concept-based interpretations in deep learning models.
problem Suboptimal final accuracy and lack of concept interaction and conditional dependencies.
method ECBMs use a set of neural networks to define joint energy, enabling concept correction and conditional dependency quantification.
result ECBMs achieve higher accuracy and richer concept interpretations compared to state-of-the-art methods.
The study identifies latent concepts from diverse observations without assuming specific models.
problem Lack of general theoretical support for concept learning.
method Develops a nonparametric framework for identifying latent concepts from multiple classes of observations.
result Correctness guarantees for concept identification without parametric assumptions.
This research generates synthetic data streams for handling concept drifts and novel classes.
problem Handling concept drifts and novel classes in dynamic data streams.
method Synthetic data stream generation for both concept drifts and novel classes.
result Demonstrates the effectiveness of unsupervised drift detectors in open set recognition.
Paper translates train track concepts to cluster algebras for pseudo-Anosov mapping classes.
problem Understanding pseudo-Anosov mapping classes on surfaces.
method Using Goncharov--Shen's potential function, the paper translates train track concepts into cluster algebra language.
result Proves sign stability of general pseudo-Anosov mapping classes.
How many bits of information are revealed by a learning algorithm for a concept class of VC-dimension d? Previous works have shown that even for d=1 the amount of information may be unbounded (tend to ∞ with the universe size). Can it be that all concepts in the class require leaking a large amount of inform…
Paper analyzes iterative learning for concept classes and learns half-spaces.
problem Learning concept classes efficiently with iterative learners.
method Analyzes various settings of iterative learning and provides a constructive algorithm for half-spaces.
result Constructive iterative algorithm for learning half-spaces from informant.
Online class imbalance learning constitutes a new problem and an emerging research topic that focusses on the challenges of online learning under class imbalance and concept drift. Class imbalance deals with data streams that have very skewed distributions while concept drift deals with changes in the class imbalance s…
Extends PAC learning theory to handle partial concepts with special properties.
problem Traditional PAC learning theory cannot handle tasks with special data properties.
method Introduces partial concepts and new PAC learning framework.
result Partial concept classes cannot be captured by traditional PAC theory.
Recurrent Neural Networks (RNNs) are among the most popular models in sequential data analysis. Yet, in the foundational PAC learning language, what concept class can it learn? Moreover, how can the same recurrent unit simultaneously learn functions from different input tokens to different output tokens, without affect…
DeepStreamCE detects new classes in streaming deep neural networks.
problem Detecting new classes in deep neural networks in a streaming environment.
method Uses autoencoder and MCOD stream-based clustering for real-time concept evolution detection.
result DeepStreamCE outperforms OpenMax in identifying concept evolution.
Evaluating, explaining, and visualizing high-level concepts in generative models, such as variational autoencoders (VAEs), is challenging in part due to a lack of known prediction classes that are required to generate saliency maps in supervised learning. While saliency maps may help identify relevant features (e.g., p…
Combines neural networks and expert rules for concept-based learning.
problem Extending concept-based learning with machine learning models.
method Form constraints for joint probability distribution and represent feasible set as a convex polytope.
result Neural networks can be trained to satisfy expert rules without violating them.
In this work we study the quantitative relation between the recursive teaching dimension (RTD) and the VC dimension (VCD) of concept classes of finite sizes. The RTD of a concept class C⊆{0,1}n, introduced by Zilles et al. (2011), is a combinatorial complexity measure characterized by the worst…
Multi-Class Incremental Learning (MCIL) aims to learn new concepts by incrementally updating a model trained on previous concepts. However, there is an inherent trade-off to effectively learning new concepts without catastrophic forgetting of previous ones. To alleviate this issue, it has been proposed to keep around a…
Concept modulation models unify identifiability and extrapolation in conditional latent variable models.
problem Reliable generalization in conditional latent variable models
method Concept modulation models (CMMs) with structure AoΛoCoX result Lifts identifiability to conditional settings and controls extrapolation through attribute potentials.
A key aspect of automating predictive machine learning entails the capability of properly triggering the update of the trained model. To this aim, suitable automatic solutions to self-assess the prediction quality and the data distribution drift between the original training set and the new data have to be devised. In …
CAVs reveal latent concept distributions, but are vulnerable to adversarial attacks.
problem Understanding latent concept encodings in AI models.
method Probabilistic perspective on CAVs, deriving mean and covariance.
result CAVs can be adversarially manipulated, highlighting a vulnerability.
Boosting improves accuracy with fewer calls to weak learners for certain concept classes.
problem Improving accuracy of learning algorithms with limited weak learner calls.
method Combines boosting and list-decodable codes to achieve better performance for specific concept classes.
result A new boosting algorithm that achieves strong learning with fewer calls to weak learners and additional samples.
Humans have an impressive ability to reason about new concepts and experiences from just a single example. In particular, humans have an ability for one-shot generalization: an ability to encounter a new concept, understand its structure, and then be able to generate compelling alternative variations of the concept. We…
Study learning and refutation in non-interactive LDP, showing sample complexity equivalence.
problem Characterize sample complexity for learning and refutation in non-interactive LDP.
method Characterize sample complexity for agnostic PAC learning in non-interactive LDP protocols.
result Optimal sample complexity for any concept class is captured by the approximate γ2~norm of a natural matrix associated with the class. Skew algebroid is a natural generalization of the concept of Lie algebroid. In this paper, for a skew algebroid E, its modular class mod(E) is defined in the classical as well as in the supergeometric formulation. It is proved that there is a homogeneous nowhere-vanishing 1-density on E* which is invariant with respect…
This work proves DP learnability implies online learnability for general classification tasks.
problem Link between differential privacy and online learning for general classification tasks.
method Establishes Ramsey-type theorems for trees to prove DP learnability implies online learnability.
result DP learnability implies online learnability for general classification tasks.
Purpose of the Conference article, intended for a wider audience, is to introduce concepts and techniques used by Bronislaw Wajnryb and the author in order to show the diffeomorphism of certain elementary algebraic surfaces, called ABC surfaces, which are not deformation equivalent. In the first part are recalled the c…
New algebraic structures on manifolds generalize supergeometry concepts.
problem Developing algebraic structures for non-commutative manifolds.
method Introducing ρ-commutative manifolds, Q-manifolds, and modular classes. result Generalized modular classes for non-commutative spaces.
New algorithm learns multiclass concepts with finite Littlestone dimension.
problem Agnostic online multiclass classification in adversarial settings.
method Multiplicative weights algorithm with experts based on subsequences.
result Proves agnostic learnability if and only if Littlestone dimension is finite.
New concept of partial law invariance connects decision theory and financial risk management.
problem Connecting decision theory and financial risk management under uncertainty.
method Characterizing partially law-invariant coherent risk measures via a novel representation formula.
result Strong partial law invariance bridges the gap between existing risk measure representations.
Improved agnostic learning time via Gaussian surface area analysis.
problem Learning polynomial threshold functions under Gaussian marginals.
method Improvement of polynomial degree required for approximation.
result Near optimal bounds on agnostic learning complexity.
Teaching models for concept learning infer concepts from observations.
problem Teaching models infer concepts from observations.
method Teaching models use MAP- and MLE-learners to infer concepts.
result The teaching dimension of a concept class can be bounded and computed in polynomial time.
Learn conditional averages in PAC framework for better predictions.
problem Learning average labels over neighborhoods in unknown concept class.
method Characterization of learnability using combinatorial parameters.
result Complete characterization and sample complexity bounds.
New neural operators learn structured patterns efficiently.
problem Learning and representing complex, structured patterns in data.
method Sparse autoencoder neural operators (SAE-NOs) parameterize concepts as functions, enabling efficient and structured representation.
result SAE-FNOs learn localized patterns and generalize across different scales and discretizations.
SketchEmbedNet learns image representations from sketches, useful for few-shot learning.
problem Learning image representations from sketches for few-shot learning.
method Training a model to produce sketches of images, focusing on informative embeddings.
result Model produces informative embeddings of novel images, classes, and datasets.
Generalizes meander diagrams to virtual knots and introduces new invariants.
problem Classifying and comparing virtual knots.
method Generalization of meander diagrams to virtual knots, introduction of virtual k-arc crossing numbers. result Classes of meander and semimeander diagrams are universal for virtual knots.
Formal models of learning from teachers need to respect certain criteria to avoid collusion. The most commonly accepted notion of collusion-freeness was proposed by Goldman and Mathias (1996), and various teaching models obeying their criterion have been studied. For each model M and each concept class C,…
Contradiction graphs reveal VC dimension threshold.
problem Determining VC dimension of concept classes.
method Study contradiction graphs of binary concept classes.
result Single contradiction graph Gm(H) determines VC dimension. Like many problems in AI in their general form, supervised learning is computationally intractable. We hypothesize that an important reason humans can learn highly complex and varied concepts, in spite of the computational difficulty, is that they benefit tremendously from experienced and insightful teachers. This pape…
Proposes a topological framework to study modular invariants and related concepts.
problem Exploring modular invariants and related concepts in topological quantum field theory.
method Topological paradigm in alterfold topological quantum field theory.
result Establishes a novel integral identity for modular invariance across multiple Morita contexts.
High-dimensional neural network manifolds misalign with human perception, causing adversarial examples.
problem Adversarial attacks fool neural networks, but their origin is unclear.
method Defined and analyzed a network's perceptual manifold (PM) for a class concept.
result Neural network PMs have orders of magnitude higher dimensions than natural human concepts, suggesting exponential misalignment.
Method makes non-interpretable models more intervenable.
problem Making non-interpretable models more understandable and controllable.
method Intervenability formalization and fine-tuning of black-box models.
result Fine-tuned black-box models are more intervenable and often better-calibrated.
We study the problem of explaining a rich class of behavioral properties of deep neural networks. Distinctively, our influence-directed explanations approach this problem by peering inside the network to identify neurons with high influence on a quantity and distribution of interest, using an axiomatically-justified in…
New algorithm learns efficiently with a simple 'yes/no' oracle.
problem Can efficient learning be achieved with a simpler oracle than ERM?
method Developed an oracle that returns 'yes' or 'no' for realizable datasets.
result Learnability is possible with a polynomial price in VC dimension.
New findings on null measurability in symmetrization interface of VC learning.
problem Null measurability issues in symmetrization interface of VC learning.
method Formalized in Lean 4, using Choquet capacitability and patching properties.
result Null-measurable bad event not Borel measurable, separating regularity levels.
For a Lie algebroid, divergences chosen in a classical way lead to a uniquely defined homology theory. They define also, in a natural way, modular classes of certain Lie algebroid morphisms. This approach, applied for the anchor map, recovers the concept of modular class due to S. Evans, J.-H. Lu, and A. Weinstein.
In this paper, the concept of balanced manifolds is generalized to reduced complex spaces: the class B and balanced spaces. Compared with the case of Kahlerian, the class B is similar to the Fujiki class C and the balanced space is similar to the Kahler space. Some properties about these complex spaces are obtained, an…
We introduce the anisotropic tensor calculus, which is a way of handling with tensors that depend on the direction remaining always in the same class. This means that the derivative of an anisotropic tensor is a tensor of the same type. As an application, we show how to define derivations using anisotropic linear conne…
This paper analyzes the variability of Concept Activation Vectors (CAVs).
problem The variability of CAVs in explaining AI models.
method Theoretical analysis and experiments on real-life datasets to quantify CAVs variability.
result The variance of CAVs decreases as 1/N, where N is the number of random examples.
Hybrid model learns novel handwritten characters better than neural or symbolic models alone.
problem Generating novel yet structured concepts.
method Neuro-symbolic model combining neural networks and probabilistic programs.
result Hybrid model outperforms alternative models in learning and generalizing novel handwritten characters.