Search Results for author:
Found 10 papers, 4 papers with code
Smooth ECE: Principled Reliability Diagrams via Kernel Smoothing
We show that a simple modification fixes both constructions: first smooth the observations using an RBF kernel, then compute the Expected Calibration Error (ECE) of this smoothed function.
When Does Optimizing a Proper Loss Yield Calibration?
Optimizing proper loss functions is popularly believed to yield predictors with good calibration properties; the intuition being that for such losses, the global optimum is to predict the ground-truth probabilities, which is indeed calibrated.
Loss Minimization Yields Multicalibration for Large Neural Networks
We show that minimizing the squared loss over all neural nets of size $n$ implies multicalibration for all but a bounded number of unlucky values of $n$.
A Unifying Theory of Distance from Calibration
We study the fundamental question of how to define and measure the distance from calibration for probabilistic predictors.
What You See is What You Get: Principled Deep Learning via Distributional Generalization
In contrast, we show that Differentially-Private (DP) training provably ensures the high-level WYSIWYG property, which we quantify using a notion of distributional generalization.
The Generic Holdout: Preventing False-Discoveries in Adaptive Data Science
In this work, we propose a new framework for adaptive science which exponentially improves on this number of queries under a restricted yet scientifically relevant setting, where the goal of the scientist is to find a single (or a few) true hypotheses about the universe based on the samples.
Optimal streaming and tracking distinct elements with high probability
The distinct elements problem is one of the fundamental problems in streaming algorithms --- given a stream of integers in the range $\{1,\ldots, n\}$, we wish to provide a $(1+\varepsilon)$ approximation to the number of distinct elements in the input.
Data Structures and Algorithms
Predicting Positive and Negative Links with Noisy Queries: Theory & Practice
The {\em edge sign prediction problem} aims to predict whether an interaction between a pair of nodes will be positive or negative.
ADAGIO: Fast Data-aware Near-Isometric Linear Embeddings
We verify experimentally the efficiency of our method on numerous real-world datasets, where we find that our method ($<$10 secs) is more than 3\, 000$\times$ faster than the state-of-the-art method \cite{hedge2015} ($>$9 hours) on medium scale datasets with 60\, 000 data points in 784 dimensions.
An improved analysis of the ER-SpUD dictionary learning algorithm
Then, given some small number $p$ of samples, i. e.\ columns of $Y$, the goal is to learn the dictionary $A$ up to small error, as well as $X$.
