96 citations · 298 across the 54 of their papers we have counts for
15 papers · 1 filter
Lifting Finite-Sample Gaussian Approximation of Wide Neural Networks to Functional Convergence
Luis Wiederhold, Nil Ayday, Debarghya Ghoshdastidar
Randomly initialised neural networks of width converge to Gaussian processes as , with two types of quantitative bounds characterising this convergence. Finite-sa…
An Analysis of Self-supervised Pre-training with Dependent Samples
Maximilian Fleissner, Debarghya Ghoshdastidar, Samory Kpotufe
Self-supervised learning relies on so-called data augmentations of unlabeled datapoints --- for example, masking random pixels in an image --- that should leave the…
Gaussian Process Limit Reveals Structural Benefits of Graph Transformers
Nil Ayday, Lingchu Yang, Debarghya Ghoshdastidar
Graph transformers are the state-of-the-art for learning from graph-structured data and are empirically known to avoid several pitfalls of message-passing architectures. However, t…
Nonparametric Kernel Clustering with Bandit Feedback
Victor Thuot, Sebastian Vogt, Debarghya Ghoshdastidar +1
Clustering with bandit feedback refers to the problem of partitioning the arms of an -armed stochastic bandit according to their underlying distributions, based on noisy observa…
Exact Generalisation Error Exposes Benchmarks Skew Graph Neural Networks Success (or Failure)
Nil Ayday, Mahalakshmi Sabanayagam, Debarghya Ghoshdastidar
Graph Neural Networks (GNNs) have become the standard method for learning from networks across fields ranging from biology to social systems, yet a principled understanding of what…
Impact of Bottleneck Layers and Skip Connections on the Generalization of Linear Denoising Autoencoders
Jonghyun Ham, Maximilian Fleissner, Debarghya Ghoshdastidar
Modern deep neural networks exhibit strong generalization even in highly overparameterized regimes. Significant progress has been made to understand this phenomenon in the context…
Tight PAC-Bayesian Risk Certificates for Contrastive Learning
Anna Van Elst, Debarghya Ghoshdastidar
Contrastive representation learning is a modern paradigm for learning representations of unlabeled data via augmentations -- precisely, contrastive models learn to embed semantical…
A Consistent Estimator for Confounding Strength
Luca Rendsburg, Leena Chennuru Vankadara, Debarghya Ghoshdastidar +1
Regression on observational data can fail to capture a causal relationship in the presence of unobserved confounding. Confounding strength measures this mismatch, but estimating it…
Interpolation and Regularization for Causal Learning
Leena Chennuru Vankadara, Luca Rendsburg, Ulrike von Luxburg +1
We study the problem of learning causal models from observational data through the lens of interpolation and its counterpart -- regularization. A large volume of recent theoretical…
Causal Forecasting:Generalization Bounds for Autoregressive Models
Leena Chennuru Vankadara, Philipp Michael Faller, Michaela Hardt +3
Despite the increasing relevance of forecasting methods, causal implications of these algorithms remain largely unexplored. This is concerning considering that, even under simplify…
On the optimality of kernels for high-dimensional clustering
Leena Chennuru Vankadara, Debarghya Ghoshdastidar
This paper studies the optimality of kernel methods in high-dimensional data clustering. Recent works have studied the large sample performance of kernel clustering in the high-dim…
Practical methods for graph two-sample testing
Debarghya Ghoshdastidar, Ulrike von Luxburg
Hypothesis testing for graphs has been an important tool in applied research fields for more than two decades, and still remains a challenging problem as one often needs to draw in…
Foundations of Comparison-Based Hierarchical Clustering
Debarghya Ghoshdastidar, Michaël Perrot, Ulrike von Luxburg
We address the classical problem of hierarchical clustering, but in a framework where one does not have access to a representation of the objects or their pairwise similarities. In…
Comparison Based Nearest Neighbor Search
Siavash Haghiri, Debarghya Ghoshdastidar, Ulrike von Luxburg
We consider machine learning in a comparison-based setting where we are given a set of points in a metric space, but we have no access to the actual distances between the points. I…
Consistency of Spectral Hypergraph Partitioning under Planted Partition Model
Debarghya Ghoshdastidar, Ambedkar Dukkipati
Hypergraph partitioning lies at the heart of a number of problems in machine learning and network sciences. Many algorithms for hypergraph partitioning have been proposed that exte…