Search Results for author:
Found 255 papers, 146 papers with code
Infrastructure Quality Assessment in Africa using Satellite Imagery and Deep Learning
Our best models predict infrastructure quality with AUROC scores of 0. 881 on Electricity, 0. 862 on Sewerage, 0. 739 on Piped Water, and 0. 786 on Roads using Landsat 8.
Amortized Inference Regularization
In this paper, we leverage the fact that VAEs rely on amortized inference and propose techniques for amortized inference regularization (AIR) that control the smoothness of the inference model.
Variational Rejection Sampling
Learning latent variable models with stochastic variational inference is challenging when the approximate posterior is far from the true posterior, due to high variance in the gradient estimates.
Best arm identification in multi-armed bandits with delayed feedback
We propose a generalization of the best arm identification problem in stochastic multi-armed bandits (MAB) to the setting where every pull of an arm is associated with delayed feedback.
Shape optimization in laminar flow with a label-guided variational autoencoder
Computational design optimization in fluid dynamics usually requires to solve non-linear partial differential equations numerically.
Deterministic Policy Optimization by Combining Pathwise and Score Function Estimators for Discrete Action Spaces
Our method is applicable to both discrete and continuous action spaces, when competing pathwise methods are limited to the latter.
Neural Variational Inference and Learning in Undirected Graphical Models
Many problems in machine learning are naturally expressed in the language of undirected graphical models.
Hierarchical Modeling of Seed Variety Yields and Decision Making for Future Planting Plans
Eradicating hunger and malnutrition is a key development goal of the 21st century.
Poverty Prediction with Public Landsat 7 Satellite Imagery and Machine Learning
Obtaining detailed and reliable data about local economic livelihoods in developing countries is expensive, and data are consequently scarce.
Fast Amortized Inference and Learning in Log-linear Models with Randomly Perturbed Nearest Neighbor Search
Inference in log-linear models scales linearly in the size of output space in the worst-case.
On the Limits of Learning Representations with Label-Based Supervision
Advances in neural network based classifiers have transformed automatic feature learning from a pipe dream of stronger AI to a routine and expected property of practical systems.
Estimating Uncertainty Online Against an Adversary
Assessing uncertainty is an important step towards ensuring the safety and reliability of machine learning systems.
Solving Marginal MAP Problems with NP Oracles and Parity Constraints
Arising from many applications at the intersection of decision making and machine learning, Marginal Maximum A Posteriori (Marginal MAP) Problems unify the two main classes of inference, namely maximization (optimization) and marginal inference (counting), and are believed to have higher complexity than both of them.
Label-Free Supervision of Neural Networks with Physics and Domain Knowledge
In many machine learning applications, labeled data is scarce and obtaining more labels is expensive.
Variable Elimination in the Fourier Domain
The ability to represent complex high dimensional probability distributions in a compact form is one of the key insights in the field of graphical models.
Model-Free Imitation Learning with Policy Optimization
In imitation learning, an agent learns how to behave in an environment with an unknown cost function by mimicking expert demonstrations.
Tight Variational Bounds via Random Projections and I-Projections
We show that information projections can be combined with random projections to obtain provable guarantees on the quality of the approximation obtained, regardless of the complexity of the original model.
Pattern Decomposition with Complex Combinatorial Constraints: Application to Materials Discovery
Identifying important components or factors in large amounts of noisy data is a key problem in machine learning and data mining.
Optimization With Parity Constraints: From Binary Codes to Discrete Integration
Many probabilistic inference tasks involve summations over exponentially large sets.
Improved Training with Curriculum GANs
We demonstrate that this strategy is key to obtaining state-of-the-art results in image generation.
Learning to Interpret Satellite Images Using Wikipedia
Despite recent progress in computer vision, fine-grained interpretation of satellite images remains challenging because of a lack of labeled training data.
Adaptive Concentration Inequalities for Sequential Decision Problems
A key challenge in sequential decision problems is to determine how many samples are needed for an agent to make reliable decisions with good probabilistic guarantees.
Variational Bayes on Monte Carlo Steroids
We provide a new approach for learning latent variable models based on optimizing our new bounds on the log-likelihood.
Embed and Project: Discrete Sampling with Universal Hashing
We consider the problem of sampling from a probability distribution defined over a high-dimensional discrete set, specified for instance by a graphical model.
Density Propagation and Improved Bounds on the Partition Function
Given a probabilistic graphical model, its density of states is a function that, for any likelihood value, gives the number of configurations with that probability.
Accelerated Adaptive Markov Chain for Partition Function Computation
We propose a novel Adaptive Markov Chain Monte Carlo algorithm to compute the partition function.
Learning Hierarchical Features from Deep Generative Models
In this paper, we prove that hierarchical latent variable models do not take advantage of the hierarchical structure when trained with existing variational methods, and provide some limitations on the kind of features existing models can learn.
LSH Softmax: Sub-Linear Learning and Inference of the Softmax Layer in Deep Architectures
In this work, we present LSH Softmax, a method to perform sub-linear learning and inference of the softmax layer in the deep learning setting.
Adversarial Examples for Natural Language Classification Problems
Modern machine learning algorithms are often susceptible to adversarial examples — maliciously crafted inputs that are undetectable by humans but that fool the algorithm into producing undesirable behavior.
The Information-Autoencoding Family: A Lagrangian Perspective on Latent Variable Generative Modeling
A variety of learning objectives have been recently proposed for training generative models.
Uncertainty Autoencoders: Learning Compressed Representations via Variational Information Maximization
We treat the low-dimensional projections as noisy latent representations of an autoencoder and directly learn both the acquisition (i. e., encoding) and amortized recovery (i. e., decoding) procedures.
Training Variational Autoencoders with Buffered Stochastic Variational Inference
The recognition network in deep latent variable models such as variational autoencoders (VAEs) relies on amortized inference for efficient posterior approximation that can scale up to large datasets.
Semi-Supervised Multitask Learning on Multispectral Satellite Images Using Wasserstein Generative Adversarial Networks (GANs) for Predicting Poverty
Obtaining reliable data describing local poverty metrics at a granularity that is informative to policy-makers requires expensive and logistically difficult surveys, particularly in the developing world.
Distributed generation of privacy preserving data with user customization
Distributed devices such as mobile phones can produce and store large amounts of data that can enhance machine learning models; however, this data may contain private information specific to the data owner that prevents the release of the data.
Predicting Economic Development using Geolocated Wikipedia Articles
Progress on the UN Sustainable Development Goals (SDGs) is hampered by a persistent lack of data regarding key social, environmental, and economic indicators, particularly in developing countries.
Mapping Missing Population in Rural India: A Deep Learning Approach with Satellite Imagery
Millions of people worldwide are absent from their country's census.
Learning Neural PDE Solvers with Convergence Guarantees
Partial differential equations (PDEs) are widely used across the physical and computational sciences.
Unsupervised Out-of-Distribution Detection with Batch Normalization
Based on this insight, we propose to exploit in-batch dependencies for OoD detection.
Out-of-Distribution Detection
Out of Distribution (OOD) Detection
Approximating Human Judgment of Generated Image Quality
Generative models have made immense progress in recent years, particularly in their ability to generate high quality images.
Generating Interpretable Poverty Maps using Object Detection in Satellite Images
Accurate local-level poverty measurement is an essential task for governments and humanitarian organizations to track the progress towards improving livelihoods and distribute scarce resources.
Farmland Parcel Delineation Using Spatio-temporal Convolutional Networks
Using satellite imaging can be a scalable and cost effective manner to perform the task of farm parcel delineation to collect this valuable data.
Efficient Poverty Mapping using Deep Reinforcement Learning
The combination of high-resolution satellite imagery and machine learning have proven useful in many sustainability-related tasks, including poverty prediction, infrastructure measurement, and forest monitoring.
Individual Calibration with Randomized Forecasting
We show that calibration for individual samples is possible in the regression setup if the predictions are randomized, i. e. outputting randomized credible intervals.
A Framework for Sample Efficient Interval Estimation with Control Variates
We consider the problem of estimating confidence intervals for the mean of a random variable, where the goal is to produce the smallest possible interval for a given number of samples.
Unsupervised Calibration under Covariate Shift
A probabilistic model is said to be calibrated if its predicted probabilities match the corresponding empirical frequencies.
Multi-label Contrastive Predictive Coding
We demonstrate that the proposed approach is able to lead to better mutual information estimation, gain empirical improvements in unsupervised representation learning, and beat a current state-of-the-art knowledge distillation method over 10 out of 13 tasks.
Privacy Preserving Recalibration under Domain Shift
In an extensive empirical study, we find that our algorithm improves calibration on domain-shift benchmarks under the constraints of differential privacy.
Non-Markovian Predictive Coding For Planning In Latent Space
High-dimensional observations are a major challenge in the application of model-based reinforcement learning (MBRL) to real-world environments.
H-divergence: A Decision-Theoretic Discrepancy Measure for Two Sample Tests
Based on ideas from decision theory, we investigate a new class of discrepancies that are based on the optimal decision loss.
Understanding Classifiers with Generative Models
Although deep neural networks are effective on supervised learning tasks, they have been shown to be brittle.
Understanding Classifier Mistakes with Generative Models
From this observation, we develop a detection criteria for samples on which a classifier is likely to fail at test time.
Imitation with Neural Density Models
We propose a new framework for Imitation Learning (IL) via density estimation of the expert's occupancy measure followed by Maximum Occupancy Entropy Reinforcement Learning (RL) using the density as a reward.
Autoregressive Score Matching
To increase flexibility, we propose autoregressive conditional score models (AR-CSM) where we parameterize the joint distribution in terms of the derivatives of univariate log-conditionals (scores), which need not be normalized.
Right Decisions from Wrong Predictions: A Mechanism Design Alternative to Individual Calibration
Decision makers often need to rely on imperfect probabilistic forecasts.
Efficient Conditional Pre-training for Transfer Learning
Finally, we improve standard ImageNet pre-training by 1-3% by tuning available models on our subsets and pre-training on a dataset filtered from a larger scale dataset.
Privacy-Constrained Policies via Mutual Information Regularized Policy Gradients
We consider the task of training a policy that maximizes reward while minimizing disclosure of certain sensitive state variables through the actions.
Neural Network Compression for Noisy Storage Devices
Compression and efficient storage of neural network (NN) parameters is critical for applications that run on resource-constrained devices.
Local Calibration: Metrics and Recalibration
In this work, we propose the local calibration error (LCE) to span the gap between average and individual reliability.
Improved Autoregressive Modeling with Distribution Smoothing
While autoregressive models excel at image compression, their sample quality is often lacking.
Improving Compositionality of Neural Networks by Decoding Representations to Inputs
In traditional software programs, it is easy to trace program logic from variables back to input, apply assertion statements to block erroneous behavior, and compose programs together.
Multi-Agent Imitation Learning with Copulas
Multi-agent imitation learning aims to train multiple agents to perform tasks from demonstrations by learning a mapping between observations and actions, which is essential for understanding physical, social, and team-play systems.
Calibrating Predictions to Decisions: A Novel Approach to Multi-Class Calibration
In this work, we introduce a new notion -- \emph{decision calibration} -- that requires the predicted distribution and true distribution to be ``indistinguishable'' to a set of downstream decision-makers.
Equivariant Neural Network for Factor Graphs
Several indices used in a factor graph data structure can be permuted without changing the underlying probability distribution.
Sphere2Vec: Self-Supervised Location Representation Learning on Spherical Surfaces
Location encoding is valuable for a multitude of tasks where both the absolute positions and local contexts (image, text, and other types of metadata) of spatial objects are needed for accurate predictions.
Provably Calibrated Regression Under Distribution Drift
Accurate uncertainty quantification is a key building block of trustworthy machine learning systems.
An Experimental Design Perspective on Exploration in Reinforcement Learning
In particular, we leverage ideas from Bayesian optimal experimental design to guide the selection of state-action queries for efficient learning.
Comparing Distributions by Measuring Differences that Affect Decision Making
Measuring the discrepancy between two probability distributions is a fundamental problem in machine learning and statistics.
H-Entropy Search: Generalizing Bayesian Optimization with a Decision-theoretic Uncertainty Measure
For special cases of the loss and design space, we develop gradient-based methods to efficiently optimize our proposed family of acquisition functions, and demonstrate that the resulting BO procedure shows strong empirical performance on a diverse set of optimization tasks.
Mind Your Bits and Errors: Prioritizing the Bits that Matter in Variational Autoencoders
In this work, we consider the task of image generative modeling with variational autoencoders and posit that the nature of high-dimensional image data distributions poses an intrinsic challenge.
Pseudo-Spherical Contrastive Divergence
Energy-based models (EBMs) offer flexible distribution parametrization.
Estimating High Order Gradients of the Data Distribution by Denoising
By leveraging Tweedie's formula on higher order moments, we generalize denoising score matching to estimate higher order derivatives.
Reliable Decisions with Threshold Calibration
We propose a stronger notion of calibration called threshold calibration, which is exactly the condition required to ensure that decision loss is predicted accurately for threshold decisions.
NECST: Neural Joint Source-Channel Coding
For reliable transmission across a noisy communication channel, classical results from information theory show that it is asymptotically optimal to separate out the source and channel coding processes.
Bias Correction of Learned Generative Models via Likelihood-free Importance Weighting
A standard technique to correct this bias is by importance weighting samples from the model by the likelihood ratio under the model and true distributions.
Towards Certified Defense for Unrestricted Adversarial Attacks
Our defense draws inspiration from differential privacy, and is based on intentionally adding noise to the classifier's outputs to limit the attacker's knowledge about the parameters.
Cross Domain Imitation Learning
Informally, CDIL is the process of learning how to perform a task optimally, given demonstrations of the task in a distinct domain.
Hybrid Mutual Information Lower-bound Estimators for Representation Learning
We illustrate that with one set of representations, the hybrid approach is able to achieve good performance on multiple downstream tasks such as classification, reconstruction, and generation.
Noisy Neural Network Compression for Analog Storage Devices
Efficient compression and storage of neural network (NN) parameters is critical for resource-constrained, downstream machine learning applications.
A Unified Framework for Multi-distribution Density Ratio Estimation
Binary density ratio estimation (DRE), the problem of estimating the ratio $p_1/p_2$ given their empirical samples, provides the foundation for many state-of-the-art machine learning algorithms such as contrastive representation learning and covariate shift adaptation.
Quantifying and Understanding Adversarial Examples in Discrete Input Spaces
In this work, we seek to understand and extend adversarial examples across domains in which inputs are discrete, particularly across new domains, such as computational biology.
Conditional Imitation Learning for Multi-Agent Games
In this work, we study the problem of conditional multi-agent imitation learning, where we have access to joint trajectory demonstrations at training time, and we must interact with and adapt to new partners at test time.
Tracking Urbanization in Developing Regions with Remote Sensing Spatial-Temporal Super-Resolution
Automated tracking of urban development in areas where construction information is not available became possible with recent advancements in machine learning and remote sensing.
Self-Similarity Priors: Neural Collages as Differentiable Fractal Representations
We investigate how to leverage the representations produced by Neural Collages in various tasks, including data compression and generation.
Modular Conformal Calibration
We introduce a versatile class of algorithms for recalibration in regression that we call Modular Conformal Calibration (MCC).
SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery
Unsupervised pre-training methods for large vision models have shown to enhance performance on downstream supervised tasks.
Multipoint-BAX: A New Approach for Efficiently Tuning Particle Accelerator Emittance via Virtual Objectives
Traditional black-box optimizers such as Bayesian optimization are slow and inefficient when dealing with such objectives as they must acquire the full series of measurements, but return only the emittance, with each query.
ButterflyFlow: Building Invertible Layers with Butterfly Matrices
Normalizing flows model complex probability distributions using maps obtained by composing invertible layers.
Generalizing Bayesian Optimization with Decision-theoretic Entropies
Bayesian optimization (BO) is a popular method for efficiently inferring optima of an expensive black-box function via a sequence of queries.
LMPriors: Pre-Trained Language Models as Task-Specific Priors
Particularly in low-data regimes, an outstanding challenge in machine learning is developing principled techniques for augmenting our models with suitable priors.
Concrete Score Matching: Generalized Score Matching for Discrete Data
To this end, we propose an analogous score function called the "Concrete score", a generalization of the (Stein) score for discrete settings.
Building Coverage Estimation with Low-resolution Remote Sensing Imagery
In this paper, we propose a method for estimating building coverage using only publicly available low-resolution satellite imagery that is more frequently updated.
Offline Imitation Learning with Suboptimal Demonstrations via Relaxed Distribution Matching
In this case, a well-known issue is the distribution shift between the learned policy and the behavior policy that collects the offline data.
Ideal Abstractions for Decision-Focused Learning
We present a methodology for formulating simplifying abstractions in machine learning systems by identifying and harnessing the utility structure of decisions.
MERMAIDE: Learning to Align Learners using Model-Based Meta-Learning
We then show that our model-based meta-learning approach is cost-effective in intervening on bandit agents with unseen explore-exploit strategies.
MUDiff: Unified Diffusion for Complete Molecule Generation
Our model is a promising approach for designing stable and diverse molecules and can be applied to a wide range of tasks in molecular modeling.
CSP: Self-Supervised Contrastive Spatial Pre-Training for Geospatial-Visual Representations
To directly leverage the abundant geospatial information associated with images in pre-training, fine-tuning, and inference stages, we present Contrastive Spatial Pre-Training (CSP), a self-supervised learning framework for geo-tagged images.
On the Equivalence of Consistency-Type Models: Consistency Models, Consistent Diffusion Models, and Fokker-Planck Regularization
The emergence of various notions of ``consistency'' in diffusion models has garnered considerable attention and helped achieve improved sample quality, likelihood estimation, and accelerated sampling.
SequenceMatch: Imitation Learning for Autoregressive Sequence Modelling with Backtracking
This allows us to minimize a variety of divergences between the distribution of sequences generated by an autoregressive model and sequences from a dataset, including divergences with weight on OOD generated sequences.
Sphere2Vec: A General-Purpose Location Representation Learning over a Spherical Surface for Large-Scale Geospatial Predictions
So when applied to large-scale real-world GPS coordinate datasets, which require distance metric learning on the spherical surface, both types of models can fail due to the map projection distortion problem (2D) and the spherical-to-Euclidean distance approximation error (3D).
SSIF: Learning Continuous Image Representation for Spatial-Spectral Super-Resolution
Existing digital sensors capture images at fixed spatial and spectral resolutions (e. g., RGB, multispectral, and hyperspectral images), and each combination requires bespoke machine learning models.
The Role of Linguistic Priors in Measuring Compositional Generalization of Vision-Language Models
Compositionality is a common property in many modalities including natural languages and images, but the compositional generalization of multi-modal models is not well-understood.
Generative Fractional Diffusion Models
We generalize the continuous time framework for score-based generative models from an underlying Brownian motion (BM) to an approximation of fractional Brownian motion (FBM).
Diffusion Model Alignment Using Direct Preference Optimization
Large language models (LLMs) are fine-tuned using human comparison data with Reinforcement Learning from Human Feedback (RLHF) methods to make them better aligned with users' preferences.
Manifold Preserving Guided Diffusion
Despite the recent advancements, conditional image generation still faces challenges of cost, generalizability, and the need for task-specific training.
DiffusionSat: A Generative Foundation Model for Satellite Imagery
Our method outperforms previous state-of-the-art methods for satellite image generation and is the first large-scale $\textit{generative}$ foundation model for satellite imagery.
Equivariant Graph Neural Operator for Modeling 3D Dynamics
Modeling the complex three-dimensional (3D) dynamics of relational systems is an important problem in the natural sciences, with applications ranging from molecular simulations to particle mechanics.
Segment Any Change
Visual foundation models have achieved remarkable results in zero-shot image classification and segmentation, but zero-shot change detection remains an open problem.
Mechanistic Design and Scaling of Hybrid Architectures
The development of deep learning architectures is a resource-demanding process, due to a vast design space, long prototyping times, and high compute costs associated with at-scale model training and evaluation.
Disentangling Length from Quality in Direct Preference Optimization
A number of approaches have been developed to control those biases in the classical RLHF literature, but the problem remains relatively under-explored for Direct Alignment Algorithms such as Direct Preference Optimization (DPO).
On the Scalability of Diffusion-based Text-to-Image Generation
On the data scaling side, we show the quality and diversity of the training set matters more than simply dataset size.
LISA: Learning Interpretable Skill Abstractions from Language
Learning policies that effectively utilize language instructions in complex, multi-task environments is an important problem in sequential decision-making.
HarvestNet: A Dataset for Detecting Smallholder Farming Activity Using Harvest Piles and Remote Sensing
Small farms contribute to a large share of the productive land in developing countries.
Approximate Inference via Weighted Rademacher Complexity
Rademacher complexity is often used to characterize the learnability of a hypothesis class and is known to be related to the class size.
Approximating the Permanent by Sampling from Adaptive Partitions
Computing the permanent of a non-negative matrix is a core problem with practical applications ranging from target tracking to statistical thermodynamics.
Calibration by Distribution Matching: Trainable Kernel Calibration Metrics
Calibration ensures that probabilistic forecasts meaningfully capture uncertainty by requiring that predicted probabilities align with empirical frequencies.
Transfer Learning from Deep Features for Remote Sensing and Poverty Mapping
We train a fully convolutional CNN model to predict nighttime lights from daytime imagery, simultaneously learning features that are useful for poverty prediction.
Differentiable Antithetic Sampling for Variance Reduction in Stochastic Variational Inference
Stochastic optimization techniques are standard in variational inference algorithms.
Imitation Learning by Estimating Expertise of Demonstrators
In this work, we show that unsupervised learning over demonstrator expertise can lead to a consistent boost in the performance of imitation learning algorithms.
Generative Modeling Helps Weak Supervision (and Vice Versa)
The model outperforms baseline weak supervision label models on a number of multiclass image classification datasets, improves the quality of generated images, and further improves end-model performance through data augmentation with synthetic samples.
Streamlining Variational Inference for Constraint Satisfaction Problems
Several algorithms for solving constraint satisfaction problems are based on survey propagation, a variational inference scheme used to obtain approximate marginal probability estimates for variable assignments.
Towards General-Purpose Representation Learning of Polygonal Geometries
For the spatial domain approach, we propose ResNet1D, a 1D CNN-based polygon encoder, which uses circular padding to achieve loop origin invariance on simple polygons.
Equivariant Flow Matching with Hybrid Probability Transport
The generation of 3D molecules requires simultaneously deciding the categorical features~(atom types) and continuous features~(atom coordinates).
Uncertainty Quantification for Forward and Inverse Problems of PDEs via Latent Global Evolution
Deep learning-based surrogate models have demonstrated remarkable advantages over classical solvers in terms of speed, often achieving speedups of 10 to 1000 times over traditional partial differential equation (PDE) solvers.
Bias Correction of Learned Generative Models using Likelihood-Free Importance Weighting
A standard technique to correct this bias is importance sampling, where samples from the model are weighted by the likelihood ratio under model and true distributions.
Deep Latent State Space Models for Time-Series Generation
Methods based on ordinary differential equations (ODEs) are widely used to build generative models of time-series.
Learning to Interpret Satellite Images in Global Scale Using Wikipedia
Despite recent progress in computer vision, finegrained interpretation of satellite images remains challenging because of a lack of labeled training data.
Experience Replay with Likelihood-free Importance Weights
The use of past experiences to accelerate temporal difference (TD) learning of value functions, or experience replay, is a key component in deep reinforcement learning.
Efficient Learning of Generative Models via Finite-Difference Score Matching
Several machine learning applications involve the optimization of higher-order derivatives (e. g., gradients of gradients) during training, which can be expensive in respect to memory and computation even with automatic differentiation.
Density Ratio Estimation via Infinitesimal Classification
We then estimate the instantaneous rate of change of the bridge distributions indexed by time (the "time score") -- a quantity defined analogously to data (Stein) scores -- with a novel time score matching objective.
Training and Inference on Any-Order Autoregressive Models the Right Way
Conditional inference on arbitrary subsets of variables is a core problem in probabilistic inference with important applications such as masked language modeling and image inpainting.
GeoLLM: Extracting Geospatial Knowledge from Large Language Models
With GeoLLM, we observe that GPT-3. 5 outperforms Llama 2 and RoBERTa by 19% and 51% respectively, suggesting that the performance of our method scales well with the size of the model and its pretraining dataset.
Large Language Models are Geographically Biased
Initially, we demonstrate that LLMs are capable of making accurate zero-shot geospatial predictions in the form of ratings that show strong monotonic correlation with ground truth (Spearman's $\rho$ of up to 0. 89).
HyperSPNs: Compact and Expressive Probabilistic Circuits
Probabilistic circuits (PCs) are a family of generative models which allows for the computation of exact likelihoods and marginals of its probability distributions.
Adversarial Constraint Learning for Structured Prediction
Constraint-based learning reduces the burden of collecting labels by having users specify general properties of structured outputs, such as constraints imposed by physical laws.
Meta-Amortized Variational Inference and Learning
Despite the recent success in probabilistic modeling and their applications, generative models trained using traditional inference techniques struggle to adapt to new distributions, even when the target distribution may be closely related to the ones seen during training.
Bridging the Gap Between $f$-GANs and Wasserstein GANs
Generative adversarial networks (GANs) have enjoyed much success in learning high-dimensional distributions.
Bridging the Gap Between f-GANs and Wasserstein GANs
Generative adversarial networks (GANs) variants approximately minimize divergences between the model and the data distribution using a discriminator.
On the Critical Role of Conventions in Adaptive Human-AI Collaboration
Humans can quickly adapt to new partners in collaborative tasks (e. g. playing basketball), because they understand which fundamental skills of the task (e. g. how to dribble, how to shoot) carry over across new partners.
Probabilistic Circuits for Variational Inference in Discrete Graphical Models
Inference in discrete graphical models with variational methods is difficult because of the inability to re-parameterize gradients of the Evidence Lower Bound (ELBO).
FP-Diffusion: Improving Score-based Diffusion Models by Enforcing the Underlying Score Fokker-Planck Equation
Score-based generative models (SGMs) learn a family of noise-conditional score functions corresponding to the data density perturbed with increasingly large amounts of noise.
Preference Fine-Tuning of LLMs Should Leverage Suboptimal, On-Policy Data
Our main finding is that, in general, approaches that use on-policy sampling or attempt to push down the likelihood on certain responses (i. e., employ a "negative gradient") outperform offline and maximum likelihood objectives.
Predictive Coding for Locally-Linear Control
High-dimensional observations and unknown dynamics are major challenges when applying optimal control to many real-world decision making tasks.
Long Horizon Temperature Scaling
LHTS is compatible with all likelihood-based models, and optimizes for the long horizon likelihood of samples.
Fair Generative Modeling via Weak Supervision
Real-world datasets are often biased with respect to key demographic factors such as race and gender.
Predicting Livelihood Indicators from Community-Generated Street-Level Imagery
Major decisions from governments and other large organizations rely on measurements of the populace's well-being, but making such measurements at a broad scale is expensive and thus infrequent in much of the developing world.
Featurized Density Ratio Estimation
Density ratio estimation serves as an important technique in the unsupervised machine learning toolbox.
Transform Once: Efficient Operator Learning in Frequency Domain
Instead, this work introduces a blueprint for frequency domain learning through a single transform: transform once (T1).
GibbsDDRM: A Partially Collapsed Gibbs Sampler for Solving Blind Inverse Problems with Denoising Diffusion Restoration
Pre-trained diffusion models have been successfully used as priors in a variety of linear inverse problems, where the goal is to reconstruct a signal from noisy linear measurements.
Temporal FiLM: Capturing Long-Range Sequence Dependencies with Feature-Wise Modulations
Learning representations that accurately capture long-range dependencies in sequential inputs -- including text, audio, and genomic data -- is a key problem in deep learning.
Ranked #2 on
Audio Super-Resolution
on Voice Bank corpus (VCTK)
(using extra training data)
Belief Propagation Neural Networks
Learned neural solvers have successfully been used to solve combinatorial optimization and decision problems.
MADiff: Offline Multi-agent Learning with Diffusion Models
To the best of our knowledge, MADiff is the first diffusion-based multi-agent offline RL framework, which behaves as both a decentralized policy and a centralized controller.
Boosted Generative Models
We propose a novel approach for using unsupervised boosting to create an ensemble of generative models, where models are trained in sequence to correct earlier mistakes.
Gaussianization Flows
Iterative Gaussianization is a fixed-point iteration procedure that can transform any continuous random vector into a Gaussian one.
An Experimental Design Perspective on Model-Based Reinforcement Learning
In particular, we leverage ideas from Bayesian optimal experimental design to guide the selection of state-action queries for efficient learning.
Exploration via Planning for Information about the Optimal Trajectory
In this work, we develop a method that allows us to plan for exploration while taking both the task and the current knowledge about the dynamics into account.
PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples
Adversarial perturbations of normal images are usually imperceptible to humans, but they can seriously confuse state-of-the-art machine learning models.
Learning Controllable Fair Representations
Learning data representations that are transferable and are fair with respect to certain protected attributes is crucial to reducing unfair decisions while preserving the utility of the data.
Domain Adaptive Imitation Learning
We formalize the Domain Adaptive Imitation Learning (DAIL) problem, which is a unified framework for imitation learning in the presence of viewpoint, embodiment, and dynamics mismatch.
Anytime Sampling for Autoregressive Models via Ordered Autoencoding
Experimentally, we demonstrate in several image and audio generation tasks that sample quality degrades gracefully as we reduce the computational budget for sampling.
Negative Data Augmentation
Empirically, models trained with our method achieve improved conditional/unconditional image generation along with improved anomaly detection capabilities.
Ranked #6 on
Image Generation
on CIFAR-100
Accelerating Feedforward Computation via Parallel Nonlinear Equation Solving
Feedforward computation, such as evaluating a neural network or sampling from an autoregressive model, is ubiquitous in machine learning.
BCD Nets: Scalable Variational Approaches for Bayesian Causal Discovery
We propose Bayesian Causal Discovery Nets (BCD Nets), a variational inference framework for estimating a distribution over DAGs characterizing a linear-Gaussian SEM.
Bias and Generalization in Deep Generative Models: An Empirical Study
In high dimensional settings, density estimation algorithms rely crucially on their inductive bias.
IS-COUNT: Large-scale Object Counting from Satellite Images with Covariate-based Importance Sampling
We show empirically that the proposed framework achieves strong performance on estimating the number of buildings in the United States and Africa, cars in Kenya, brick kilns in Bangladesh, and swimming pools in the U. S., while requiring as few as 0. 01% of satellite images compared to an exhaustive approach.
The Information Autoencoding Family: A Lagrangian Perspective on Latent Variable Generative Models
A large number of objectives have been proposed to train latent variable generative models.
Spatial-Temporal Super-Resolution of Satellite Imagery via Conditional Pixel Synthesis
High-resolution satellite imagery has proven useful for a broad range of tasks, including measurement of global human population, local economic livelihoods, and biodiversity, among many others.
Accelerating Natural Gradient with Higher-Order Invariance
An appealing property of the natural gradient is that it is invariant to arbitrary differentiable reparameterizations of the model.
Temporal Predictive Coding For Model-Based Planning In Latent Space
High-dimensional observations are a major challenge in the application of model-based reinforcement learning (MBRL) to real-world environments.
AlignFlow: Cycle Consistent Learning from Multiple Domains via Normalizing Flows
Given datasets from multiple domains, a key challenge is to efficiently exploit these data sources for modeling a target domain.
Training Deep Energy-Based Models with f-Divergence Minimization
Experimental results demonstrate the superiority of f-EBM over contrastive divergence, as well as the benefits of training EBMs using f-divergences other than KL.
Accurate Uncertainties for Deep Learning Using Calibrated Regression
Methods for reasoning under uncertainty are a key building block of accurate and reliable machine learning systems.
Geography-Aware Self-Supervised Learning
Contrastive learning methods have significantly narrowed the gap between supervised and unsupervised learning on computer vision tasks.
Ranked #5 on
Semantic Segmentation
on SpaceNet 1
(using extra training data)
Evaluating the Disentanglement of Deep Generative Models through Manifold Topology
Learning disentangled representations is regarded as a fundamental task for improving the generalization, robustness, and interpretability of generative models.
MintNet: Building Invertible Neural Networks with Masked Convolutions
To demonstrate their flexibility, we show that our invertible neural networks are competitive with ResNets on MNIST and CIFAR-10 classification.
Ranked #4 on
Image Generation
on MNIST
Cross-Modal Contextualized Diffusion Models for Text-Guided Visual Generation and Editing
To address this issue, we propose a novel and general contextualized diffusion model (ContextDiff) by incorporating the cross-modal context encompassing interactions and alignments between text condition and visual sample into forward and reverse processes.
Improving Self-Supervised Learning by Characterizing Idealized Representations
For non-contrastive learning, we use our framework to derive a simple and novel objective.
A General Recipe for Likelihood-free Bayesian Optimization
To extend BO to a broader class of models and utilities, we propose likelihood-free BO (LFBO), an approach based on likelihood-free inference.
Bayesian Algorithm Execution: Estimating Computable Properties of Black-box Functions Using Mutual Information
Given such an $\mathcal{A}$, and a prior distribution over $f$, we refer to the problem of inferring the output of $\mathcal{A}$ using $T$ evaluations as Bayesian Algorithm Execution (BAX).
Neural Joint Source-Channel Coding
For reliable transmission across a noisy communication channel, classical results from information theory show that it is asymptotically optimal to separate out the source and channel coding processes.
Reparameterizable Subset Sampling via Continuous Relaxations
Many machine learning tasks require sampling a subset of items from a collection based on a parameterized distribution.
End-to-End Diffusion Latent Optimization Improves Classifier Guidance
Classifier guidance -- using the gradients of an image classifier to steer the generations of a diffusion model -- has the potential to dramatically expand the creative control over image generation and editing.
JPEG Artifact Correction using Denoising Diffusion Restoration Models
Diffusion models can be used as learned priors for solving various inverse problems.
Modeling Sparse Deviations for Compressed Sensing using Generative Models
In compressed sensing, a small number of linear measurements can be used to reconstruct an unknown signal.
Diversity can be Transferred: Output Diversification for White- and Black-box Attacks
Adversarial attacks often involve random perturbations of the inputs drawn from uniform or Gaussian distributions, e. g., to initialize optimization-based white-box attacks or generate update directions in black-box attacks.
GlueGen: Plug and Play Multi-modal Encoders for X-to-image Generation
Empirical results show that GlueNet can be trained efficiently and enables various capabilities beyond previous state-of-the-art models: 1) multilingual language models such as XLM-Roberta can be aligned with existing T2I models, allowing for the generation of high-quality images from captions beyond English; 2) GlueNet can align multi-modal encoders such as AudioCLIP with the Stable Diffusion model, enabling sound-to-image generation; 3) it can also upgrade the current text encoder of the latent diffusion model for challenging case generation.
Calibrated Model-Based Deep Reinforcement Learning
Estimates of predictive uncertainty are important for accurate model-based planning and reinforcement learning.
Model-based Reinforcement Learning
reinforcement-learning
+1
Graphite: Iterative Generative Modeling of Graphs
Graphs are a fundamental abstraction for modeling relational data.
Ranked #8 on
Link Prediction
on Citeseer
PiRank: Scalable Learning To Rank via Differentiable Sorting
A key challenge with machine learning approaches for ranking is the gap between the performance metrics of interest and the surrogate loss functions that can be optimized with gradient-based methods.
Constructing Unrestricted Adversarial Examples with Generative Models
Then, conditioned on a desired class, we search over the AC-GAN latent space to find images that are likely under the generative model and are misclassified by a target classifier.
Cloud Removal in Satellite Images Using Spatiotemporal Generative Networks
In contrast, we cast the problem of cloud removal as a conditional image synthesis challenge, and we propose a trainable spatiotemporal generator network (STGAN) to remove clouds.
Ranked #6 on
Cloud Removal
on SEN12MS-CR-TS
GEO-Bench: Toward Foundation Models for Earth Monitoring
Recent progress in self-supervision has shown that pre-training large neural networks on vast amounts of unsupervised data can lead to substantial increases in generalization to downstream tasks.
Understanding the Limitations of Variational Mutual Information Estimators
Variational approaches based on neural networks are showing promise for estimating mutual information (MI) between high dimensional variables.
Extreme Q-Learning: MaxEnt RL without Entropy
Using EVT, we derive our \emph{Extreme Q-Learning} framework and consequently online and, for the first time, offline MaxEnt Q-learning algorithms, that do not explicitly require access to a policy or its entropy.
Learning When and Where to Zoom with Deep Reinforcement Learning
While high resolution images contain semantically more useful information than their lower resolution counterparts, processing them is computationally more expensive, and in some applications, e. g. remote sensing, they can be much more expensive to acquire.
Meta-Inverse Reinforcement Learning with Probabilistic Context Variables
Critically, our model can infer rewards for new, structurally-similar tasks from a single demonstration.
Ranked #1 on
MuJoCo Games
on Sawyer Pusher
Efficient Object Detection in Large Images using Deep Reinforcement Learning
Traditionally, an object detector is applied to every part of the scene of interest, and its accuracy and computational cost increases with higher resolution images.
A Theory of Usable Information Under Computational Constraints
We propose a new framework for reasoning about information in complex systems.
Semi-supervised Deep Kernel Learning: Regression with Unlabeled Data by Minimizing Predictive Variance
Large amounts of labeled data are typically required to train deep learning models.
HIVE: Harnessing Human Feedback for Instructional Visual Editing
Incorporating human feedback has been shown to be crucial to align text generated by large language models to human preferences.
Rapid identification of pathogenic bacteria using Raman spectroscopy and deep learning
By amassing the largest known dataset of bacterial Raman spectra, we are able to apply state-of-the-art deep learning approaches to identify 30 of the most common bacterial pathogens from noisy Raman spectra, achieving antibiotic treatment identification accuracies of 99. 0$\pm$0. 1%.
