147 citations · 696 across the 57 of their papers we have counts for
25 papers · 1 filter
Bridging Human and LLM Judgments: Understanding and Narrowing the Gap
Felipe Maia Polo, Xinhe Wang, Mikhail Yurochkin +3
Large language models are increasingly used as judges (LLM-as-a-judge) to evaluate model outputs at scale, but their assessments often diverge systematically from human judgments.…
Sloth: scaling laws for LLM skills to predict multi-benchmark performance across families
Felipe Maia Polo, Seamus Somerstep, Leshem Choshen +2
Scaling laws for large language models (LLMs) predict model performance based on parameters like size and training data. However, differences in training configurations and data pr…
Distributional Preference Alignment of LLMs via Optimal Transport
Igor Melnyk, Youssef Mroueh, Brian Belgodere +6
Current LLM alignment techniques use pairwise human preferences at a sample level, and as such, they do not imply an alignment on the distributional level. We propose in this paper…
Asymmetry in Low-Rank Adapters of Foundation Models
Jiacheng Zhu, Kristjan Greenewald, Kimia Nadjahi +6
Parameter-efficient fine-tuning optimizes large, pre-trained foundation models by updating a subset of parameters; in this class, Low-Rank Adaptation (LoRA) is particularly effecti…
Uncertainty Quantification via Stable Distribution Propagation
Felix Petersen, Aashwin Mishra, Hilde Kuehne +3
We propose a new approach for propagating stable probability distributions through neural networks. Our method is based on local linearization, which we show to be an optimal appro…
GeRA: Label-Efficient Geometrically Regularized Alignment
Dustin Klebe, Tal Shnitzer, Mikhail Yurochkin +2
Pretrained unimodal encoders incorporate rich semantic information into embedding space structures. To be similarly informative, multi-modal encoders typically require massive amou…
Risk Aware Benchmarking of Large Language Models
Apoorva Nitsure, Youssef Mroueh, Mattia Rigotti +6
We propose a distributional framework for benchmarking socio-technical risks of foundation models with quantified statistical significance. Our approach hinges on a new statistical…
Fusing Models with Complementary Expertise
Hongyi Wang, Felipe Maia Polo, Yuekai Sun +3
Training AI models that generalize across tasks and domains has long been among the open problems driving AI research. The emergence of Foundation Models made it easier to obtain e…
Simple Disentanglement of Style and Content in Visual Representations
Lilian Ngweta, Subha Maity, Alex Gittens +2
Learning visual representations with interpretable features, i.e., disentangled representations, remains a challenging problem. Existing methods demonstrate some success but are ha…
Outlier-Robust Group Inference via Gradient Space Clustering
Yuchen Zeng, Kristjan Greenewald, Kangwook Lee +2
Traditional machine learning models focus on achieving good performance on the overall training distribution, but they often underperform on minority groups. Existing methods can i…
How does overparametrization affect performance on minority groups?
Subha Maity, Saptarshi Roy, Songkai Xue +2
The benefits of overparameterization for the overall performance of modern machine learning (ML) models are well known. However, the effect of overparameterization at a more granul…
Understanding new tasks through the lens of training data via exponential tilting
Subha Maity, Mikhail Yurochkin, Moulinath Banerjee +1
Deploying machine learning models to new tasks is a major challenge despite the large size of the modern training datasets. However, it is conceivable that the training data can be…
Rewiring with Positional Encodings for Graph Neural Networks
Rickard Brüel-Gabrielsson, Mikhail Yurochkin, Justin Solomon
Several recent works use positional encodings to extend the receptive fields of graph neural network (GNN) layers equipped with attention mechanisms. These techniques, however, ext…
Learning Proximal Operators to Discover Multiple Optima
Lingxiao Li, Noam Aigerman, Vladimir G. Kim +4
Finding multiple solutions of non-convex optimization problems is a ubiquitous yet challenging task. Most past algorithms either apply single-solution optimization methods from mul…
On sensitivity of meta-learning to support data
Mayank Agarwal, Mikhail Yurochkin, Yuekai Sun
Meta-learning algorithms are widely used for few-shot learning. For example, image recognition systems that readily adapt to unseen classes after seeing only a few labeled examples…
k-Mixup Regularization for Deep Learning via Optimal Transport
Kristjan Greenewald, Anming Gu, Mikhail Yurochkin +2
Mixup is a popular regularization technique for training deep neural networks that improves generalization and increases robustness to certain distribution shifts. It perturbs inpu…
Individually Fair Gradient Boosting
Alexander Vargo, Fan Zhang, Mikhail Yurochkin +1
We consider the task of enforcing individual fairness in gradient boosting. Gradient boosting is a popular method for machine learning from tabular data, which arise often in appli…
Online Semi-Supervised Learning with Bandit Feedback
Sohini Upadhyay, Mikhail Yurochkin, Mayank Agarwal +2
We formulate a new problem at the intersectionof semi-supervised learning and contextual bandits,motivated by several applications including clini-cal trials and ad recommendations…
Continuous Regularized Wasserstein Barycenters
Lingxiao Li, Aude Genevay, Mikhail Yurochkin +1
Wasserstein barycenters provide a geometrically meaningful way to aggregate probability distributions, built on the theory of optimal transport. They are difficult to compute in pr…
IBM Federated Learning: an Enterprise Framework White Paper V0.1
Heiko Ludwig, Nathalie Baracaldo, Gegi Thomas +21
Federated Learning (FL) is an approach to conduct machine learning without centralizing training data in a single place, for reasons of privacy, confidentiality or data volume. How…
Model Fusion with Kullback--Leibler Divergence
Sebastian Claici, Mikhail Yurochkin, Soumya Ghosh +1
We propose a method to fuse posterior distributions learned from heterogeneous datasets. Our algorithm relies on a mean field assumption for both the fused model and the individual…
SenSeI: Sensitive Set Invariance for Enforcing Individual Fairness
Mikhail Yurochkin, Yuekai Sun
In this paper, we cast fair machine learning as invariant machine learning. We first formulate a version of individual fairness that enforces invariance on certain sensitive sets.…
Federated Learning with Matched Averaging
Hongyi Wang, Mikhail Yurochkin, Yuekai Sun +2
Federated learning allows edge devices to collaboratively learn a shared model while keeping the training data on device, decoupling the ability to do model training from the need…
Alleviating Label Switching with Optimal Transport
Pierre Monteiller, Sebastian Claici, Edward Chien +3
Label switching is a phenomenon arising in mixture model posterior inference that prevents one from meaningfully assessing posterior statistics using standard Monte Carlo procedure…