Greg Ver Steeg
YOU?
Author Swipe
View article: The Upside of Bias: Personalizing Long-Tail Item Recommendations with Biased Sampling
The Upside of Bias: Personalizing Long-Tail Item Recommendations with Biased Sampling Open
Recommendation systems drive user engagement across social media, streaming platforms, and e-commerce by learning from past interactions. The relevance of a recommended item depends on the quality of the user and item embeddings learned by…
View article: Local MAP Sampling for Diffusion Models
Local MAP Sampling for Diffusion Models Open
Diffusion Posterior Sampling (DPS) provides a principled Bayesian approach to inverse problems by sampling from $p(x_0 \mid y)$. However, in practice, the goal of inverse problem solving is not to cover the posterior but to recover the mos…
View article: MMG: Mutual Information Estimation via the MMSE Gap in Diffusion
MMG: Mutual Information Estimation via the MMSE Gap in Diffusion Open
Mutual information (MI) is one of the most general ways to measure relationships between random variables, but estimating this quantity for complex systems is challenging. Denoising diffusion models have recently set a new bar for density …
View article: Generating diffusion MRI scalar maps from T1-weighted images using Reversible GANs
Generating diffusion MRI scalar maps from T1-weighted images using Reversible GANs Open
Diffusion tensor imaging (DTI) provides valuable insights into brain tissue microstructure, but acquiring high-quality DTI data is time-intensive and not always feasible. To mitigate data scarcity and enhance accessibility, we investigate …
View article: Measurement-Aligned Sampling for Inverse Problem
Measurement-Aligned Sampling for Inverse Problem Open
Diffusion models provide a powerful way to incorporate complex prior information for solving inverse problems. However, existing methods struggle to correctly incorporate guidance from conflicting signals in the prior and measurement, and …
View article: Synthetic Diffusion Tensor Imaging Maps Generated by 2D and 3D Probabilistic Diffusion Models: Evaluation and Applications
Synthetic Diffusion Tensor Imaging Maps Generated by 2D and 3D Probabilistic Diffusion Models: Evaluation and Applications Open
Diffusion tensor imaging (DTI) is a key neuroimaging modality for assessing brain tissue microstructure, yet high-quality acquisitions are costly, time-intensive, and prone to artifacts. To address data scarcity and privacy concerns – and …
View article: Making Sense Of Distributed Representations With Activation Spectroscopy
Making Sense Of Distributed Representations With Activation Spectroscopy Open
In the study of neural network interpretability, there is growing evidence to suggest that relevant features are encoded across many neurons in a distributed fashion. Making sense of these distributed representations without knowledge of t…
View article: Learning Morphisms with Gauss-Newton Approximation for Growing Networks
Learning Morphisms with Gauss-Newton Approximation for Growing Networks Open
A popular method for Neural Architecture Search (NAS) is based on growing networks via small local changes to the network's architecture called network morphisms. These methods start with a small seed network and progressively grow the net…
View article: Exploring the Design Space of Diffusion Bridge Models
Exploring the Design Space of Diffusion Bridge Models Open
Diffusion bridge models and stochastic interpolants enable high-quality image-to-image (I2I) translation by creating paths between distributions in pixel space. However, the proliferation of techniques based on incompatible mathematical as…
View article: QuAILoRA: Quantization-Aware Initialization for LoRA
QuAILoRA: Quantization-Aware Initialization for LoRA Open
QLoRA reduces the memory-cost of fine-tuning a large language model (LLM) with LoRA by quantizing the base LLM. However, quantization introduces quantization errors that negatively impact model performance after fine-tuning. In this paper …
View article: Biased User History Synthesis for Personalized Long-Tail Item Recommendation
Biased User History Synthesis for Personalized Long-Tail Item Recommendation Open
View article: Comparison of Explainable AI Models for MRI-based Alzheimer’s Disease Classification
Comparison of Explainable AI Models for MRI-based Alzheimer’s Disease Classification Open
Deep learning models based on convolutional neural networks (CNNs) have been used to classify Alzheimer’s disease or infer dementia severity from 3D T1-weighted brain MRI scans. Here, we examine the value of adding occlusion sensitivity an…
View article: A federated learning architecture for secure and private neuroimaging analysis
A federated learning architecture for secure and private neuroimaging analysis Open
The amount of biomedical data continues to grow rapidly. However, collecting data from multiple sites for joint analysis remains challenging due to security, privacy, and regulatory concerns. To overcome this challenge, we use federated le…
View article: Your Diffusion Model is Secretly a Noise Classifier and Benefits from Contrastive Training
Your Diffusion Model is Secretly a Noise Classifier and Benefits from Contrastive Training Open
Diffusion models learn to denoise data and the trained denoiser is then used to generate new samples from the data distribution. In this paper, we revisit the diffusion sampling process and identify a fundamental cause of sample quality de…
View article: Comparison of deep learning architectures for predicting amyloid positivity in Alzheimer’s disease, mild cognitive impairment, and healthy aging, from T1-weighted brain structural MRI
Comparison of deep learning architectures for predicting amyloid positivity in Alzheimer’s disease, mild cognitive impairment, and healthy aging, from T1-weighted brain structural MRI Open
Abnormal β-amyloid (Aβ) accumulation in the brain is an early indicator of Alzheimer’s disease (AD) and is typically assessed through invasive procedures such as PET (positron emission tomography) or CSF (cerebrospinal fluid) assays. As ne…
View article: q-Diffusion leverages the full dimensionality of gene coexpression in single-cell transcriptomics
q-Diffusion leverages the full dimensionality of gene coexpression in single-cell transcriptomics Open
View article: Advances in AI for Protein Structure Prediction: Implications for Cancer Drug Discovery and Development
Advances in AI for Protein Structure Prediction: Implications for Cancer Drug Discovery and Development Open
Recent advancements in AI-driven technologies, particularly in protein structure prediction, are significantly reshaping the landscape of drug discovery and development. This review focuses on the question of how these technological breakt…
View article: Prompt Perturbation Consistency Learning for Robust Language Models
Prompt Perturbation Consistency Learning for Robust Language Models Open
Large language models (LLMs) have demonstrated impressive performance on a number of natural language processing tasks, such as question answering and text summarization. However, their performance on sequence labeling tasks such as intent…
View article: Interpretable Measures of Conceptual Similarity by Complexity-Constrained Descriptive Auto-Encoding
Interpretable Measures of Conceptual Similarity by Complexity-Constrained Descriptive Auto-Encoding Open
Quantifying the degree of similarity between images is a key copyright issue for image-based machine learning. In legal doctrine however, determining the degree of similarity between works requires subjective analysis, and fact-finders (ju…
View article: Asymmetric Bias in Text-to-Image Generation with Adversarial Attacks
Asymmetric Bias in Text-to-Image Generation with Adversarial Attacks Open
The widespread use of Text-to-Image (T2I) models in content generation requires careful examination of their safety, including their robustness to adversarial attacks. Despite extensive research on adversarial attacks, the reasons for thei…
View article: Interpretable Diffusion via Information Decomposition
Interpretable Diffusion via Information Decomposition Open
Denoising diffusion models enable conditional generation and density modeling of complex relationships like images and text. However, the nature of the learned relationships is opaque making it difficult to understand precisely what relati…
View article: Ensembled Prediction Intervals for Causal Outcomes Under Hidden Confounding
Ensembled Prediction Intervals for Causal Outcomes Under Hidden Confounding Open
Causal inference of exact individual treatment outcomes in the presence of hidden confounders is rarely possible. Recent work has extended prediction intervals with finite-sample guarantees to partially identifiable causal outcomes, by mea…
View article: Knowledge Enhanced Multi-Domain Recommendations in an AI Assistant Application
Knowledge Enhanced Multi-Domain Recommendations in an AI Assistant Application Open
This work explores unifying knowledge enhanced recommendation with multi-domain recommendation systems in a conversational AI assistant application. Multi-domain recommendation leverages users' interactions in previous domains to improve r…
View article: Jointly Reparametrized Multi-Layer Adaptation for Efficient and Private Tuning
Jointly Reparametrized Multi-Layer Adaptation for Efficient and Private Tuning Open
Efficient finetuning of pretrained language transformers is becoming increasingly prevalent for solving natural language processing tasks. While effective, it can still require a large number of tunable parameters. This can be a drawback f…
View article: Neural Architecture Search for Parameter-Efficient Fine-tuning of Large Pre-trained Language Models
Neural Architecture Search for Parameter-Efficient Fine-tuning of Large Pre-trained Language Models Open
Parameter-efficient tuning (PET) methods fit pre-trained language models (PLMs) to downstream tasks by either computing a small compressed update for a subset of model parameters, or appending and fine-tuning a small number of new model pa…
View article: Measuring and Mitigating Local Instability in Deep Neural Networks
Measuring and Mitigating Local Instability in Deep Neural Networks Open
Deep Neural Networks (DNNs) are becoming integral components of real world services relied upon by millions of users. Unfortunately, architects of these systems can find it difficult to ensure reliable performance as irrelevant details lik…
View article: Transferring Models Trained on Natural Images to 3D MRI via Position Encoded Slice Models
Transferring Models Trained on Natural Images to 3D MRI via Position Encoded Slice Models Open
Transfer learning has remarkably improved computer vision. These advances also promise improvements in neuroimaging, where training set sizes are often small. However, various difficulties arise in directly applying models pretrained on na…
View article: Improving Mutual Information Estimation with Annealed and Energy-Based Bounds
Improving Mutual Information Estimation with Annealed and Energy-Based Bounds Open
Mutual information (MI) is a fundamental quantity in information theory and machine learning. However, direct estimation of MI is intractable, even if the true joint probability density for the variables of interest is known, as it involve…
View article: Transferring Models Trained on Natural Images to 3D MRI via Position Encoded Slice Models
Transferring Models Trained on Natural Images to 3D MRI via Position Encoded Slice Models Open
Transfer learning has remarkably improved computer vision. These advances also promise improvements in neuroimaging, where training set sizes are often small. However, various difficulties arise in directly applying models pretrained on na…
View article: Information-Theoretic Diffusion
Information-Theoretic Diffusion Open
Denoising diffusion models have spurred significant gains in density modeling and image generation, precipitating an industrial revolution in text-guided AI art generation. We introduce a new mathematical foundation for diffusion models in…