Transformers as statisticians: Provable in-context learning with in-context algorithm selection
Neural sequence models based on the transformer architecture have demonstrated
remarkable\emph {in-context learning}(ICL) abilities, where they can perform new tasks …
remarkable\emph {in-context learning}(ICL) abilities, where they can perform new tasks …
Assessing generalization of SGD via disagreement
We empirically show that the test error of deep networks can be estimated by simply training
the same architecture on the same training set but with a different run of Stochastic Gradient …
the same architecture on the same training set but with a different run of Stochastic Gradient …
Beyond confidence: Reliable models should also consider atypicality
While most machine learning models can provide confidence in their predictions, confidence
is insufficient to understand a prediction's reliability. For instance, the model may have a low …
is insufficient to understand a prediction's reliability. For instance, the model may have a low …
Early-stopped neural networks are consistent
Z Ji, J Li, M Telgarsky - Advances in Neural Information …, 2021 - proceedings.neurips.cc
This work studies the behavior of shallow ReLU networks trained with the logistic loss via
gradient descent on binary classification data where the underlying data distribution is …
gradient descent on binary classification data where the underlying data distribution is …
On double-descent in uncertainty quantification in overparametrized models
Uncertainty quantification is a central challenge in reliable and trustworthy machine
learning. Naive measures such as last-layer scores are well-known to yield overconfident …
learning. Naive measures such as last-layer scores are well-known to yield overconfident …
T-cal: An optimal test for the calibration of predictive models
The prediction accuracy of machine learning methods is steadily increasing, but the
calibration of their uncertainty predictions poses a significant challenge. Numerous works …
calibration of their uncertainty predictions poses a significant challenge. Numerous works …
Conformal prediction sets improve human decision making
In response to everyday queries, humans explicitly signal uncertainty and offer alternative
answers when they are unsure. Machine learning models that output calibrated prediction …
answers when they are unsure. Machine learning models that output calibrated prediction …
Multi-label intent detection via contrastive task specialization of sentence encoders
Deploying task-oriented dialog ToD systems for new domains and tasks requires natural
language understanding models that are 1) resource-efficient and work under low-data …
language understanding models that are 1) resource-efficient and work under low-data …
Federated inference with reliable uncertainty quantification over wireless channels via conformal prediction
In this paper, we consider a wireless federated inference scenario in which devices and a
server share a pre-trained machine learning model. The devices communicate statistical …
server share a pre-trained machine learning model. The devices communicate statistical …
Analysis of bootstrap and subsampling in high-dimensional regularized regression
We investigate popular resampling methods for estimating the uncertainty of statistical
models, such as subsampling, bootstrap and the jackknife, and their performance in high …
models, such as subsampling, bootstrap and the jackknife, and their performance in high …