arXiv research

Characterizes sample complexity for outcome indistinguishability in machine learning.

The paper examines how to test if two learning algorithms produce similar outcomes.

Efficiently generates models resistant to falsification.

Survey on calibration in machine learning, viewing it as indistinguishability.

We show social events can be accurately predicted, but often undesirably.

We resolve the open problem of optimal sample complexity for multicalibration and deterministic predictors.

New approach tackles decision-making under predictions that shape outcomes.

The paper reviews and extends calibration concepts for classification and regression.

The paper targets optimal interventions for long-term outcomes using imputed data and policy learning.

Generative Adversarial Networks are a new family of generative models, frequently used for generating photorealistic images. The theory promises for the GAN to eventually reach an equilibrium where generator produces pictures indistinguishable for the training set. In practice, however, a range of problems frequently p…

We inject undetectable backdoors into obfuscated neural networks and language models.

New index principle shows indistinguishability of certain knot crossings.

Framework uses human judgment to distinguish algorithmically indistinguishable cases.

A new metric evaluates classification algorithms at the point of indistinguishability.

The paper formalizes robot environments using topological concepts.

We investigate the possibility of statistical evaluation of the market completeness for discrete time stock market models. It is known that the market completeness is not a robust property: small random deviations of the coefficients convert a complete market model into a incomplete one. The paper shows that market inc…

New approach shows backdoor attacks are indistinguishable from natural data features.

Transformers approximate mean-field dynamics of indistinguishable particles.

The possibility of statistical evaluation of the market completeness and incompleteness is investigated for continuous time diffusion stock market models. It is known that the market completeness is not a robust property: small random deviations of the coefficients convert a complete market model into a incomplete one.…

HappyMap improves fairness and learning across domains by generalizing multi-calibration.

Graph convolutional networks (GCNs) are a widely used method for graph representation learning. To elucidate the capabilities and limitations of GCNs, we investigate their power, as a function of their number of layers, to distinguish between different random graph models (corresponding to different class-conditional d…

Unsupervised learning models can be indistinguishable without identifiability, leading to unreliable representations.

Graphs indistinguishable by GNNs are fully characterized.

Time dilation $\frac{1}{\sqrt{1-v^2}}$ and relative velocity $v$ are observationally indistinguishable in the special theory of relativity, a duality that carries over into the general theory under Fermi coordinates along a curve (in coordinate-independent language, in the tangent Minkowski space along the curve). For …

Efficient algorithms for deleting data from machine learning models without significantly affecting performance.

Study on replicability in reinforcement learning algorithms.

Market timing is an investment technique that tries to continuously switch investment into assets forecast to have better returns. What is the likelihood of having a successful market timing strategy? With an emphasis on modeling simplicity, I calculate the feasible set of market timing portfolios using index mutual fu…

Two knots in three-space are S-equivalent if they are indistinguishable by Seifert matrices. We show that S-equivalence is generated by the doubled-delta move on knot diagrams. It follows as a corollary that a knot has trivial Alexander polynomial if and only if it can be undone by doubled-delta moves.

Differential privacy has emerged as a gold standard in privacy-preserving data analysis. A popular variant is local differential privacy, where the data holder is the trusted curator. A major barrier, however, towards a wider adoption of this model is that it offers a poor privacy-utility tradeoff. In this work, we add…

We establish tools to facilitate the computation and application of the Chekanov-Eliashberg differential graded algebra (DGA), a Legendrian-isotopy invariant of Legendrian knots in standard contact three-space. More specifically, we reformulate the DGA in terms of front projection, and introduce the characteristic alge…

Paper distinguishes causal structures under latent confounding and selection bias.

Paper proposes a GAN-based method for better next event prediction in business processes.

Graph neural networks fail to distinguish certain 3D atom configurations.

This paper introduces a method to approximate Gaussian process regression by representing the problem as a stochastic differential equation and using variational inference to approximate solutions. The approximations are compared with full GP regression and generated paths are demonstrated to be indistinguishable from …

PathNNs improve graph neural networks by distinguishing non-isomorphic graphs.

Generative models create indistinguishable adversarial objects for object detection.

AI predicts stock winners with 2.43 Sharpe ratio, but returns are highly concentrated.

Data augmentation is essential for medical research to increase the size of training datasets and achieve better results. In this work, we experiment three GAN architectures with different loss functions to generate new brain MRIs. The results show the importance of hyperparameter tuning and the use of mini-batch simil…

A new approach for instance-optimal learning that bypasses impossibility results.

Paper simplifies balancing weights by relaxing outcome assumptions.

Enhanced attacks quantify machine learning data leakage.

A method corrects bias in estimating a high-dimensional classification rule using auxiliary outcomes.

We study, using Mean Curvature Flow methods, 2+1 dimensional cosmologies with a positive cosmological constant and matter satisfying the dominant and the strong energy conditions. If the spatial slices are compact with non-positive Euler characteristic and are initially expanding everywhere, then we prove that the spat…

PO-Flow models potential and counterfactual outcomes for personalized treatment decisions.

Fuses ITRs for primary and secondary outcomes to minimize harm.

We investigate the complexity of logistic regression models which is defined by counting the number of indistinguishable distributions that the model can represent (Balasubramanian, 1997). We find that the complexity of logistic models with binary inputs does not only depend on the number of parameters but also on the …

Firms delay write-downs for adverse macroeconomic and industry outcomes but not for firm-specific issues.

Reduces variance in noisy social outcomes to improve policy evaluation and optimization.