Search Results for author:
Found 44 papers, 24 papers with code
On Implicit Regularization in $\beta$-VAEs
While the impact of variational inference (VI) on posterior inference in a fixed generative model is well-characterized, its role in regularizing a learned generative model when used in variational autoencoders (VAEs) is poorly understood.
Diffusion Self-Guidance for Controllable Image Generation
However, many aspects of an image are difficult or impossible to convey through text.
Learning a Diffusion Prior for NeRFs
Neural Radiance Fields (NeRFs) have emerged as a powerful neural 3D representation for objects and scenes derived from 2D data.
DreamBooth3D: Subject-Driven Text-to-3D Generation
We present DreamBooth3D, an approach to personalize text-to-3D generative models from as few as 3-6 casually captured images of a subject.
VeLO: Training Versatile Learned Optimizers by Scaling Up
While deep learning models have replaced hand-designed features across many domains, these models are still trained with hand-designed optimizers.
Imagen Video: High Definition Video Generation with Diffusion Models
We present Imagen Video, a text-conditional video generation system based on a cascade of video diffusion models.
Ranked #1 on
Video Generation
on LAION-400M
DreamFusion: Text-to-3D using 2D Diffusion
Using this loss in a DeepDream-like procedure, we optimize a randomly-initialized 3D model (a Neural Radiance Field, or NeRF) via gradient descent such that its 2D renderings from random angles achieve a low loss.
Zero-Shot Text-Guided Object Generation with Dream Fields
Our method, Dream Fields, can generate the geometry and color of a wide range of objects without 3D supervision.
On Density Estimation with Diffusion Models
In addition, we show that the continuous-time VLB is invariant to the noise schedule, except for the signal-to-noise ratio at its endpoints.
Autoregressive Diffusion Models
We introduce Autoregressive Diffusion Models (ARDMs), a model class encompassing and generalizing order-agnostic autoregressive models (Uria et al., 2014) and absorbing discrete diffusion (Austin et al., 2021), which we show are special cases of ARDMs under mild assumptions.
Ranked #6 on
Image Generation
on CIFAR-10
(bits/dimension metric)
Variational Diffusion Models
In addition, we show that the continuous-time VLB is invariant to the noise schedule, except for the signal-to-noise ratio at its endpoints.
Ranked #1 on
Image Generation
on CIFAR-10
(bits/dimension metric)
TaskSet: A Dataset of Optimization Tasks
We present TaskSet, a dataset of tasks for use in training and evaluating optimizers.
Overcoming barriers to the training of effective learned optimizers
In this work we focus on general-purpose learned optimizers capable of training a wide variety of problems with no user-specified hyperparameters.
Learning Energy-Based Models by Diffusion Recovery Likelihood
Inspired by recent progress on diffusion probabilistic models, we present a diffusion recovery likelihood method to tractably learn and sample from a sequence of EBMs trained on increasingly noisy versions of a dataset.
Ranked #14 on
Image Generation
on CelebA 64x64
Score-Based Generative Modeling through Stochastic Differential Equations
Combined with multiple architectural improvements, we achieve record-breaking performance for unconditional image generation on CIFAR-10 with an Inception score of 9. 89 and FID of 2. 20, a competitive likelihood of 2. 99 bits/dim, and demonstrate high fidelity generation of 1024 x 1024 images for the first time from a score-based generative model.
Ranked #11 on
Image Generation
on CIFAR-10
VIB is Half Bayes
In discriminative settings such as regression and classification there are two random variables at play, the inputs X and the targets Y.
Non-saturating GAN training as divergence minimization
Non-saturating generative adversarial network (GAN) training is widely used and has continued to obtain groundbreaking results.
Tasks, stability, architecture, and compute: Training more effective learned optimizers, and using them to train themselves
In this work we focus on general-purpose learned optimizers capable of training a wide variety of problems with no user-specified hyperparameters.
What Makes for Good Views for Contrastive Learning?
Contrastive learning between multiple views of the data has recently achieved state of the art performance in the field of self-supervised representation learning.
Ranked #2 on
Contrastive Learning
on imagenet-1k
Using a thousand optimization tasks to learn hyperparameter search strategies
We present TaskSet, a dataset of tasks for use in training and evaluating optimizers.
Regularized Autoencoders via Relaxed Injective Probability Flow
Invertible flow-based generative models are an effective method for learning to generate samples, while allowing for tractable likelihood computation and inference.
Weakly-Supervised Disentanglement Without Compromises
Third, we perform a large-scale empirical study and show that such pairs of observations are sufficient to reliably learn disentangled representations on several benchmark data sets.
On Implicit Regularization in $β$-VAEs
While the impact of variational inference (VI) on posterior inference in a fixed generative model is well-characterized, its role in regularizing a learned generative model when used in variational autoencoders (VAEs) is poorly understood.
Weakly Supervised Disentanglement with Guarantees
Learning disentangled representations that correspond to factors of variation in real-world data is critical to interpretable and human-controllable machine learning.
On Predictive Information in RNNs
Certain biological neurons demonstrate a remarkable capability to optimally compress the history of sensory inputs while being maximally informative about the future.
On Predictive Information Sub-optimality of RNNs
Certain biological neurons demonstrate a remarkable capability to optimally compress the history of sensory inputs while being maximally informative about the future.
Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentation
Deploying machine learning systems in the real world requires both high accuracy on clean data and robustness to naturally occurring corruptions.
Discrete Flows: Invertible Generative Models of Discrete Data
While normalizing flows have led to significant advances in modeling high-dimensional continuous distributions, their applicability to discrete distributions remains unknown.
Ranked #16 on
Language Modelling
on Text8
On Variational Bounds of Mutual Information
Estimating and optimizing Mutual Information (MI) is core to many problems in machine learning; however, bounding MI in high dimensions is challenging.
Preventing Posterior Collapse with delta-VAEs
Due to the phenomenon of "posterior collapse," current latent variable generative models pose a challenging design choice that either weakens the capacity of the decoder or requires augmenting the objective so it does not only maximize the likelihood of the data.
Ranked #7 on
Image Generation
on ImageNet 32x32
(bpd metric)
An information-theoretic analysis of deep latent-variable models
We present an information-theoretic framework for understanding trade-offs in unsupervised learning of deep latent-variables models using variational inference.
Fixing a Broken ELBO
Recent work in unsupervised representation learning has focused on learning deep directed latent-variable models.
Continual Learning Through Synaptic Intelligence
While deep learning has led to remarkable advances across diverse applications, it struggles in domains where the data distribution changes over the course of learning.
Improved generator objectives for GANs
We present a framework to understand GAN training as alternating density ratio estimation and approximate divergence minimization.
Survey of Expressivity in Deep Neural Networks
This quantity grows exponentially in the depth of the network, and is responsible for the depth sensitivity observed.
Unrolled Generative Adversarial Networks
We introduce a method to stabilize Generative Adversarial Networks (GANs) by defining the generator objective with respect to an unrolled optimization of the discriminator.
Categorical Reparameterization with Gumbel-Softmax
Categorical variables are a natural choice for representing discrete structure in the world.
Categorical Reparametrization with Gumbel-Softmax
Categorical variables are a natural choice for representing discrete structure in the world.
Exponential expressivity in deep neural networks through transient chaos
We combine Riemannian geometry with the mean field theory of high dimensional chaos to study the nature of signal propagation in generic, deep neural networks with random weights.
On the Expressive Power of Deep Neural Networks
We propose a new approach to the problem of neural network expressivity, which seeks to characterize how structural properties of a neural network family affect the functions it is able to compute.
Adversarially Learned Inference
We introduce the adversarially learned inference (ALI) model, which jointly learns a generation network and an inference network using an adversarial process.
The Fast Bilateral Solver
We present the bilateral solver, a novel algorithm for edge-aware smoothing that combines the flexibility and speed of simple filtering approaches with the accuracy of domain-specific optimization algorithms.
Analyzing noise in autoencoders and deep networks
Autoencoders have emerged as a useful framework for unsupervised learning of internal representations, and a wide variety of apparently conceptually disparate regularization techniques have been proposed to generate useful features.
Fast large-scale optimization by unifying stochastic gradient and quasi-Newton methods
This algorithm contrasts with earlier stochastic second order techniques that treat the Hessian of each contributing function as a noisy approximation to the full Hessian, rather than as a target for direct estimation.
