Deep Generative Models are frequently used to learn continuous representations of complex data distributions by training on a finite number of samples. For any generative model, including pre-trained foundation models with Diffusion or Transformer architectures, generation performance can significantly vary across the learned data manifold. In this paper, we study the local geometry of the learned manifold and its relationship to generation outcomes for a wide range of generative models, including DDPM, Diffusion Transformer (DiT), and Stable Diffusion 1.4. Building on the theory of continuous piecewise-linear (CPWL) generators, we characterize the local geometry in terms of three geometric descriptors - scaling ($\psi$), rank ($\nu$), and complexity/un-smoothness ($\delta$). We provide quantitative and qualitative evidence showing that for a given latent vector, the local descriptors are indicative of post-generation aesthetics, generation diversity, and memorization by the generative model. Finally, we demonstrate that by training a reward model on the 'local scaling' for Stable Diffusion, we can self-improve both generation aesthetics and diversity using geometry sensitive guidance during denoising. Website: https://imtiazhumayun.github.io/generative_geometry.

The rise of foundation models holds immense promise for advancing AI, but this progress may amplify existing risks and inequalities, leaving marginalized communities behind. In this position paper, we discuss that disparities towards marginalized communities – performance, representation, privacy, robustness, interpretability and safety – are not isolated concerns but rather interconnected elements of a cascading disparity phenomenon. We contrast foundation models with traditional models and highlight the potential for exacerbated disparity against marginalized communities. Moreover, we emphasize the unique threat of cascading impacts in foundation models, where interconnected disparities can trigger long-lasting negative consequences, specifically to the people on the margin. We define marginalized communities within the machine learning context and explore the multifaceted nature of disparities. We analyze the sources of these disparities, tracing them from data creation, training and deployment procedures to highlight the complex technical and socio-technical landscape. To mitigate the pressing crisis, we conclude with a set of calls to action to mitigate disparity at its source.

Batch normalization is a staple of computer vision models, including those employed in few-shot learning. Batch nor-malization layers in convolutional neural networks are composed of a normalization step, followed by a shift and scale of these normalized features applied via the per-channel trainable affine parameters$\gamma$and$\beta$. These affine param-eters were introduced to maintain the expressive powers of the model following normalization. While this hypothesis holds true for classification within the same domain, this work illustrates that these parameters are detrimen-tal to downstream performance on common few-shot trans-fer tasks. This effect is studied with multiple methods on well-known benchmarks such as few-shot classification on minilmageNet, cross-domain few-shot learning (CD-FSL) and META-DATASET. Experiments reveal consistent performance improvements on CNNs with affine unaccompanied batch normalization layers; particularly in large domain-shift few-shot transfer settings. As opposed to common practices in few-shot transfer learning where the affine pa-rameters are fixed during the adaptation phase, we show fine-tuning them can lead to improved performance.

Learning in nonstationary environments is one of the biggest challenges in machine learning. Nonstationarity can be caused by either task drift, i.e., the drift in the conditional distribution of labels given the input data, or the domain drift, i.e., the drift in the marginal distribution of the input data. This article aims to tackle this challenge with a modularized two-stream continual learning (CL) system, where the model is required to learn new tasks from a support stream and adapted to new domains in the query stream while maintaining previously learned knowledge. To deal with both drifts within and across the two streams, we propose a variational domain-agnostic feature replay-based approach that decouples the system into three modules: an inference module that filters the input data from the two streams into domain-agnostic representations, a generative module that facilitates the high-level knowledge transfer, and a solver module that applies the filtered and transferable knowledge to solve the queries. We demonstrate the effectiveness of our proposed approach in addressing the two fundamental scenarios and complex scenarios in two-stream CL.

We propose a hypothesis disparity regularized mutual information maximization (HDMI) approach to tackle unsupervised hypothesis transfer---as an effort towards unifying hypothesis transfer learning (HTL) and unsupervised domain adaptation (UDA)---where the knowledge from a source domain is transferred solely through hypotheses and adapted to the target domain in an unsupervised manner. In contrast to the prevalent HTL and UDA approaches that typically use a single hypothesis, HDMI employs multiple hypotheses to leverage the underlying distributions of the source and target hypotheses. To better utilize the crucial relationship among different hypotheses---as opposed to unconstrained optimization of each hypothesis independently---while adapting to the unlabeled target domain through mutual information maximization, HDMI incorporates a hypothesis disparity regularization that coordinates the target hypotheses jointly learn better target representations while preserving more transferable source knowledge with better-calibrated prediction uncertainty. HDMI achieves state-of-the-art adaptation performance on benchmark datasets for UDA in the context of HTL, without the need to access the source data during the adaptation.

We present an approach for unsupervised domain adaptation{—}with a strong focus on practical considerations of within-domain class imbalance and between-domain class distribution shift{—}from a class-conditioned domain alignment perspective. Current methods for class-conditioned domain alignment aim to explicitly minimize a loss function based on pseudo-label estimations of the target domain. However, these methods suffer from pseudo-label bias in the form of error accumulation. We propose a method that removes the need for explicit optimization of model parameters from pseudo-labels. Instead, we present a sampling-based implicit alignment approach, where the sample selection is implicitly guided by the pseudo-labels. Theoretical analysis reveals the existence of a domain-discriminator shortcut in misaligned classes, which is addressed by the proposed approach to facilitate domain-adversarial learning. Empirical results and ablation studies confirm the effectiveness of the proposed approach, especially in the presence of within-domain class imbalance and between-domain class distribution shift.

Synthesizing images from a given text description involves engaging two types of information: the content, which includes information explicitly described in the text (e.g., color, composition, etc.), and the style, which is usually not well described in the text (e.g., location, quantity, size, etc.). However, in previous works, it is typically treated as a process of generating images only from the content, i.e., without considering learning meaningful style representations. In this paper, we aim to learn two variables that are disentangled in the latent space, representing content and style respectively. We achieve this by augmenting current text-to-image synthesis frameworks with a dual adversarial inference mechanism. Through extensive experiments, we show that our model learns, in an unsupervised manner, style representations corresponding to certain meaningful information present in the image that are not well described in the text. The new framework also improves the quality of synthesized images when evaluated on Oxford-102, CUB and COCO datasets.

We consider the problem of brain tumor segmentation from magnetic resonance (MR) images. This task is most frequently tackled using machine learning methods that generalize across brains, by learning from training brain images in order to generalize to novel test brains. However this approach faces many obstacles that threaten its performance, such as the ability to properly perform multi-brain registration or brain-atlas alignment, or to extract appropriate high-dimensional features that support good generalization. These operations are both nontrivial and time-consuming, limiting the practicality of these approaches in a clinical context. In this paper, we propose to side step these issues by approaching the problem as one of within brain generalization. Specifically, we propose a semi-automatic method that segments a given brain by training and generalizing within that brain only, based on some minimum user interaction. We investigate how k nearest neighbors (kNN), arguably the simplest machine learning method available, combined with the simplest feature vector possible (raw MR signal + (x,y,z) position) can be combined into a method that is both simple, accurate and fast. Results obtained on the online BRATS dataset reveal that our method is fast and second best in terms of the complete and core test set tumor segmentation.