Unsupervised method discovers interpretable directions in GAN latent space.
problem Discovering interpretable directions in GAN latent space without supervision.
method Model-agnostic procedure to identify directions corresponding to semantic manipulations.
result Findings include directions for background removal and competitive saliency detection performance.
Enhances interpretability of linear latent spaces through automated clustering and ranking.
problem Severe interpretability issues in latent directions of PCA, ICA, CCA, and FA.
method LS-PIE framework automates clustering and ranking of latent vectors.
result Enhanced interpretability of latent vectors through LR, LS, LC, and LCON.
Automated method finds meaningful directions in neural network activations.
problem Mixed selectivity in neurons makes interpretation challenging.
method Automated quantification of interpretability and discovery of meaningful directions.
result Meaningful directions in neural network activations are more interpretable than individual neurons.
Generating high-quality and interpretable adversarial examples in the text domain is a much more daunting task than it is in the image domain. This is due partly to the discrete nature of text, partly to the problem of ensuring that the adversarial examples are still probable and interpretable, and partly to the proble…
Framework for interpreting ML models to reveal properties of real-world phenomena.
problem Lack of direct interpretability in modern ML models hinders scientific understanding.
method Developed 'property descriptors' grounded in statistical learning theory.
result Property descriptors can reveal relevant properties of joint probability distributions.
Deep neural networks (DNNs) have had many successes, but they suffer from two major issues: (1) a vulnerability to adversarial examples and (2) a tendency to elude human interpretation. Interestingly, recent empirical and theoretical evidence suggests these two seemingly disparate issues are actually connected. In part…
A new framework quantifies how model explanations influence each other.
problem Understanding how different model explanations interact and influence each other.
method Introducing the metagame, a conceptual framework for measuring second-order interaction effects of model explanations using Shapley values.
result Meta-attributions provide directional insights into how feature interactions influence model explanations.
A novel framework infers causal direction from symbolic sequences using pattern entropy.
problem Challenges in discovering causal direction from temporal symbolic data.
method Dictionary Based Pattern Entropy (DPE) framework integrating AIT and Shannon Information Theory. result Minimizing pattern level uncertainty yields a robust framework for causal discovery.
Study relaxes identification assumptions for natural direct effects in non-randomized settings.
problem Identifying causal direct effects under unmeasured confounding.
method Developed relaxed conditions for identifying natural direct effects in non-randomized settings.
result Identified natural direct effect under unmeasured confounding conditions.
FastMap-D embeds directed graphs using potential fields.
problem Embedding directed graphs in Euclidean space.
method Generalization of FastMap to handle directed graphs using a potential field and machine learning.
result FastMap-D outperforms other approaches in embedding directed graphs.
We give a homological interpretation of the coefficients of the Hilbert series for an algebra associated with a directed graph and its dual algebra. This allows us to obtain necessary conditions for Koszulity of such algebras in terms of homological properties of the graphs. We use our results to construct algebras wit…
Proposes a new model for directed graphs combining deep learning and latent variable models.
problem Graph representation learning for directed graphs.
method Deep Latent Space Model (DLSM) integrating GCN encoder and stochastic decoder with hierarchical variational auto-encoder architecture.
result Achieves state-of-the-art performance on link prediction and community detection tasks.
Enhances math problem-solving models with multi-turn preference learning.
problem Improving mathematical problem-solving capabilities of large language models.
method Introduces a multi-turn direct preference learning framework for tool-integrated mathematical reasoning tasks.
result Significant performance improvements in model accuracy on math datasets.
The paper interprets policy-gradient algorithms using continuation theory.
problem Optimizing nonconvex functions in reinforcement learning.
method Formulates policy optimization as optimization by continuation, interprets policy-gradient algorithms as implicitly optimizing deterministic policies.
result Exploration in policy-gradient algorithms is seen as computing a continuation of the return of the policy.
New framework improves interpretability of trainable prompts.
problem Improving task-specific LLM performance with soft prompts remains a black-box method.
method Developed a theoretical framework for evaluating interpretability of trainable prompts, inspired new objective functions.
result Found a fundamental trade-off between interpretability and task performance in trainable prompts.
CDFD analyzes circularity and directionality in weighted directed networks.
problem Analyzing circularity and directionality in weighted directed networks.
method CDFD framework separates flow into circular and acyclic components.
result CDFD yields a normalized circularity index capturing flow in cycles and directionality.
CaTs use DAGs with transformers to enforce causal constraints, improving neural network robustness.
problem Neural networks lack inherent causal structure respect, leading to reliability issues.
method Introducing Causal Transformers (CaTs) that operate under predefined causal constraints specified by DAGs.
result CaTs improve robustness and interpretability of neural networks under causal constraints.
The paper introduces a new method for interpreting model predictions by considering both direct and indirect effects.
problem Interpreting model predictions to understand the causes of decisions.
method Proposes a new approach to quantify feature relevance by combining different types of interpretations and measures.
result Integrates various types of interpretations and measures to provide meaningful insights into model predictions.
ARTEMIS combines deep learning and symbolic reasoning for financial predictions.
problem Lack of interpretability and economic principles in deep learning models in finance.
method Neuro-symbolic framework combining neural operators, stochastic differential equations, and symbolic distillation.
result ARTEMIS achieves state-of-the-art directional accuracy, outperforming all baselines on synthetic crash regime.
We give a direct interpretation of Neumann's combinatorial formula for the Chern-Simons invariant of a 3-manifold with a representation in PSL(2,C) whose restriction to the boundary takes values in upper triangular matrices. Our construction does not involve group homology or Bloch group but is based on the constructio…
Most recent work on interpretability of complex machine learning models has focused on estimating a posteriori explanations for previously trained models around specific predictions. Self-explaining models where interpretability plays a key role already during learning have received much less atte…
This paper addresses anisotropy in Transformer models, providing geometric insights and empirical support.
problem Anisotropy phenomenon in Transformer models, challenging their geometric interpretation.
method Derive geometric arguments and use concept-based mechanistic interpretability during training.
result Activation-derived directions capture large gradient energy and a larger share of gradient anisotropy than normal controls.
Transparency, user trust, and human comprehension are popular ethical motivations for interpretable machine learning. In support of these goals, researchers evaluate model explanation performance using humans and real world applications. This alone presents a challenge in many areas of artificial intelligence. In this …
Hidden tree Markov models allow learning distributions for tree structured data while being interpretable as nondeterministic automata. We provide a concise summary of the main approaches in literature, focusing in particular on the causality assumptions introduced by the choice of a specific tree visit direction. We w…
We introduce a new class of graphical models that generalizes Lauritzen-Wermuth-Frydenberg chain graphs by relaxing the semi-directed acyclity constraint so that only directed cycles are forbidden. Moreover, up to two edges are allowed between any pair of nodes. Specifically, we present local, pairwise and global Marko…
Interpretability of ML models improves healthcare decisions.
problem Ensuring machine learning models are understandable for healthcare users.
method Classifying interpretability into local and global approaches, and model-specific vs. model-agnostic methods.
result Examples of practical interpretability in healthcare, including prediction and treatment optimization.
In the present work, torsion energy is defined. Its law of conservation is given. It is shown that this type of energy gives rise to a repulsive force which can be used to interpret supernovae type Ia observations, and consequently the accelerating expansion of the Universe. This interpretation is a pure geometric one …
We explore the geometrical interpretation of the PCA based clustering algorithm Principal Direction Divisive Partitioning (PDDP). We give several examples where this algorithm breaks down, and suggest a new method, gap partitioning, which takes into account natural gaps in the data between clusters. Geometric features …
Survey on understanding neural networks for medical applications.
problem Black-box nature of deep neural networks hinders their use in critical applications.
method Comprehensive review of interpretability studies in neural networks.
result Interpretability research is crucial for the acceptance of neural networks in medical diagnosis.
We introduce a novel harmonic analysis for functions defined on the vertices of a strongly connected directed graph of which the random walk operator is the cornerstone. As a first step, we consider the set of eigenvectors of the random walk operator as a non-orthogonal Fourier-type basis for functions over directed gr…
With the advent of highly predictive but opaque deep learning models, it has become more important than ever to understand and explain the predictions of such models. Existing approaches define interpretability as the inverse of complexity and achieve interpretability at the cost of accuracy. This introduces a risk of …
The paper shows equivalent interpretations of Laplacian bounds in RCD spaces.
problem Equivalence of Laplacian bounds in RCD spaces.
method Analyzes different interpretations of the inequality Δf ≤ η in RCD(K,N) spaces.
result Improves generality and regularity assumptions for function f.
Interprets deep learning models for rough volatility pricing.
problem Lack of interpretability in deep learning models for financial models.
method Detailed analysis of neural network learned inverse map between rough volatility model parameters and implied volatilities.
result Provides insights into neural network outputs for rough volatility models.
New method clusters directed graphs using Koopman operators.
problem Challenges in clustering directed graphs, especially complex eigenvalues and lack of cluster definition.
method Relate graph Laplacians to transfer operators and metastable sets in stochastic systems, derive clustering algorithms for directed and time-evolving graphs.
result Clusters can be interpreted as coherent sets, useful for analyzing transport and mixing processes.
SGD benefits from a directional bias in kernel regression models.
problem Improving generalization in kernel regression models.
method Generalized directional bias property of SGD in kernel regression.
result SGD converges along the eigenvector of the largest eigenvalue of the Gram matrix.
MAGIC-Flow generates and classifies medical images with interpretability.
problem Challenges in generative modeling for medical imaging.
method Conditional multiscale normalizing flow architecture.
result MAGIC-Flow creates realistic, diverse samples and improves classification.
Decomposes bias in linear models under demographic parity constraints.
problem Understanding and quantifying bias in linear models under fairness constraints.
method Post-processing framework to decompose bias into direct and indirect components.
result Analytical characterization of how demographic parity reshapes model coefficients.
In this paper we introduce a novel family of decision lists consisting of highly interpretable models which can be learned efficiently in a greedy manner. The defining property is that all rules are oriented in the same direction. Particular examples of this family are decision lists with monotonically decreasing (or i…
A new principle minimizes residual and introduces momentum to improve PDE solution dynamics.
problem Ill-conditioning in Dirac-Frenkel residual minimization leads to non-unique parameter dynamics.
method Introduces a history variable (momentum) to select better-conditioned parameter velocities, preserving residual minimization while promoting smooth parameter evolutions.
result The approach leads to increased robustness in singular and near-singular PDE solution regimes.
New methods improve feature extraction and representation quality in supervised and unsupervised DR.
problem Statistical dependence, data diversity, contrast, and interpretability in conventional DR methods.
method Combines linear and nonlinear formulations for three new independence criteria.
result Significant improvements in contrast, accuracy, and interpretability over baselines.
Directly estimates CQC, improving interpretability and accuracy.
problem Inability to model and interpret CQC due to inversion issue.
method Direct doubly robust estimation of CQC without inversion.
result Improved estimation accuracy and interpretability.
DiMMSB models directed mixed membership networks, identifying distinct community structures.
problem Modeling directed mixed membership networks with distinct community structures.
method Directed Mixed Membership Stochastic Blockmodel (DiMMSB) with DiSP algorithm.
result DiSP algorithm is asymptotically consistent and outperforms competitors.
B-cos transforms improve neural network interpretability by aligning weights.
problem Improving interpretability of deep neural networks.
method Replacing linear transforms with B-cos transforms that promote weight-input alignment during training.
result B-cos transforms lead to highly interpretable and task-relevant linear summaries of neural network computations.
This work interprets SFA through variational inference, relaxing linearity constraints.
problem Recover non-linear SFA from variational inference.
method Probabilistic interpretation of SFA through variational inference, relaxing linearity constraints.
result Reinterprets SFA as a variational framework, allowing slowness as a regularizer to reconstruction loss.
New algorithm consistently orients eigenvectors for machine learning.
problem Inconsistent eigenvector orientation in machine learning.
method Postprocesses well-established eigen calls to create consistently oriented eigenvectors.
result Interpretable time series of training weights in machine learning models.
New algorithms for interpreting complex multivariate functions.
problem Hard interpretation of multivariate functions due to many parameters.
method Filtered tensor decompositions of derivative information.
result Nonparametric estimates of smooth decoupled functions.
Market activity scales near a constant of 0.632 in intrinsic time.
problem Understanding the stability of market scaling laws.
method Modeling market directional changes as a memoryless exponential hazard process and identifying the intrinsic time scaling constant.
result The intrinsic time scaling constant is 1−1/e=0.632. Interpreting machine learning models helps understand adversarial attacks and defenses.
problem Understanding model vulnerability to adversarial attacks.
method Model interpretation techniques to explore adversarial attacks and defenses.
result Interpretation methods can be applied to adversarial attacks and defenses.