Search Results for author:
Found 83 papers, 28 papers with code
DIMAT: Decentralized Iterative Merging-And-Training for Deep Learning Models
This DIMAT paradigm presents a new opportunity for the future decentralized learning, enhancing its adaptability to real-world with sparse and light-weight communication and computation.
Robust Concept Erasure Using Task Vectors
Finally, we show that Diverse Inversion enables us to apply a TV edit only to a subset of the model weights, enhancing the erasure capabilities while better maintaining the core functionality of the model.
Mitigating the Impact of Attribute Editing on Face Recognition
To mitigate this issue, we propose two novel techniques for local and global attribute editing.
TuneTables: Context Optimization for Scalable Prior-Data Fitted Networks
Similar to large language models, PFNs make use of pretraining and in-context learning to achieve strong performance on new tasks in a single forward pass.
Scaling TabPFN: Sketching and Feature Selection for Tabular Prior-Data Fitted Networks
Tabular classification has traditionally relied on supervised algorithms, which estimate the parameters of a prediction model using its training data.
Exploring Dataset-Scale Indicators of Data Quality
Modern computer vision foundation models are trained on massive amounts of data, incurring large economic and environmental costs.
ArcheType: A Novel Framework for Open-Source Column Type Annotation using Large Language Models
We introduce ArcheType, a simple, practical method for context sampling, prompt serialization, model querying, and label remapping, which enables large language models to solve CTA problems in a fully zero-shot manner.
Ranked #1 on
Column Type Annotation
on WDC SOTAB
(Weighted F1 metric)
PriViT: Vision Transformers for Fast Private Inference
The Vision Transformer (ViT) architecture has emerged as the backbone of choice for state-of-the-art deep models for computer vision applications.
On the Fine-Grained Hardness of Inverting Generative Models
This paper aims to provide a fine-grained view of the landscape of computational hardness for this problem.
Towards Foundational AI Models for Additive Manufacturing: Language Models for G-Code Debugging, Manipulation, and Comprehension
3D printing or additive manufacturing is a revolutionary technology that enables the creation of physical objects from digital models.
Distributionally Robust Classification on a Data Budget
To our knowledge, this is the first result showing (near) state-of-the-art distributional robustness on limited data budgets.
Circumventing Concept Erasure Methods For Text-to-Image Generative Models
Text-to-image generative models can produce photo-realistic images for an extremely broad range of concepts, and their usage has proliferated widely among the general public.
Identity-Preserving Aging of Face Images via Latent Diffusion Models
The performance of automated face recognition systems is inevitably impacted by the facial aging process.
Vision-Language Models can Identify Distracted Driver Behavior from Naturalistic Videos
Our results show that this framework offers state-of-the-art performance on zero-shot transfer and video-based CLIP for predicting the driver's state on two public datasets.
ZeroForge: Feedforward Text-to-Shape Without 3D Supervision
Current state-of-the-art methods for text-to-shape generation either require supervised training using a labeled dataset of pre-defined 3D shapes, or perform expensive inference-time optimization of implicit neural representations.
When Do Neural Nets Outperform Boosted Trees on Tabular Data?
To this end, we conduct the largest tabular data analysis to date, comparing 19 algorithms across 176 datasets, and we find that the 'NN vs. GBDT' debate is overemphasized: for a surprisingly high number of datasets, either the performance difference between GBDTs and NNs is negligible, or light hyperparameter tuning on a GBDT is more important than choosing between NNs and GBDTs.
LiT Tuned Models for Efficient Species Detection
Recent advances in training vision-language models have demonstrated unprecedented robustness and transfer learning effectiveness; however, standard computer vision datasets are image-only, and therefore not well adapted to such training methods.
Implicit Regularization for Group Sparsity
We study the implicit regularization of gradient descent towards structured sparsity via a novel neural reparameterization, which we call a diagonally grouped linear neural network.
Pathfinding Neural Cellular Automata
As a step toward developing such networks, we hand-code and learn models for Breadth-First Search (BFS), i. e. shortest path finding, using the unified architectural framework of Neural Cellular Automata, which are iterative neural networks with equal-size inputs and outputs.
Neural PDE Solvers for Irregular Domains
Neural network-based approaches for solving partial differential equations (PDEs) have recently received special attention.
Active Learning for Single Neuron Models with Lipschitz Non-Linearities
Namely, we can collect samples via statistical \emph{leverage score sampling}, which has been shown to be near-optimal in other active learning scenarios.
Caption supervision enables robust learners
Vision language (VL) models like CLIP are robust to natural distribution shifts, in part because CLIP learns on unstructured data using a technique called caption supervision; the model inteprets image-linked texts as ground-truth labels.
GOTCHA: Real-Time Video Deepfake Detection via Challenge-Response
With the rise of AI-enabled Real-Time Deepfakes (RTDFs), the integrity of online video interactions has become a growing concern.
Distributed Online Non-convex Optimization with Composite Regret
To address these two issues, we propose a novel composite regret with a new network regret-based metric to evaluate distributed online optimization algorithms.
On The Computational Complexity of Self-Attention
Transformer architectures have led to remarkable progress in many state-of-art applications.
Revisiting Self-Distillation
We first show that even with a highly accurate teacher, self-distillation allows a student to surpass the teacher in all cases.
A Meta-Analysis of Distributionally-Robust Models
State-of-the-art image classifiers trained on massive datasets (such as ImageNet) have been shown to be vulnerable to a range of both intentional and incidental distribution shifts.
Smooth-Reduce: Leveraging Patches for Improved Certified Robustness
Randomized smoothing (RS) has been shown to be a fast, scalable technique for certifying the robustness of deep neural network classifiers.
Selective Network Linearization for Efficient Private Inference
To reduce PI latency we propose a gradient-based algorithm that selectively linearizes ReLUs while maintaining prediction accuracy.
MDPGT: Momentum-based Decentralized Policy Gradient Tracking
We propose a novel policy gradient method for multi-agent reinforcement learning, which leverages two different variance-reduction techniques and does not require large batches over iterations.
Multi-agent Reinforcement Learning
Policy Gradient Methods
+3
Adversarial Token Attacks on Vision Transformers
Vision transformers rely on a patch token based self attention mechanism, in contrast to convolutional networks.
NeuFENet: Neural Finite Element Solutions with Theoretical Bounds for Parametric PDEs
We consider a mesh-based approach for training a neural network to produce field predictions of solutions to parametric partial differential equations (PDEs).
Differentiable Spline Approximations
Overall, we show that leveraging this redesigned Jacobian in the form of a differentiable "layer" in predictive models leads to improved performance in diverse applications such as image segmentation, 3D point cloud reconstruction, and finite element analysis.
Implicit Sparse Regularization: The Impact of Depth and Early Stopping
In this paper, we study the implicit bias of gradient descent for sparse regression.
Sphynx: ReLU-Efficient Network Design for Private Inference
The emergence of deep learning has been accompanied by privacy concerns surrounding users' data and service providers' models.
Provably Convergent Algorithms for Solving Inverse Problems Using Generative Models
The traditional approach of hand-crafting priors (such as sparsity) for solving inverse problems is slowly being replaced by the use of richer learned priors (such as those modeled by deep generative networks).
NURBS-Diff: A Differentiable Programming Module for NURBS
These derivatives are used to define an approximate Jacobian used for performing the "backward" evaluation to train the deep learning models.
Distributed Multigrid Neural Solvers on Megavoxel Domains
We specifically consider neural solvers for the generalized 3D Poisson equation over megavoxel domains.
Cross-Gradient Aggregation for Decentralized Learning from Non-IID data
Inspired by ideas from continual learning, we propose Cross-Gradient Aggregation (CGA), a novel decentralized learning algorithm where (i) each agent aggregates cross-gradient information, i. e., derivatives of its model with respect to its neighbors' datasets, and (ii) updates its model using a projected gradient based on quadratic programming (QP).
Provable Compressed Sensing with Generative Priors via Langevin Dynamics
Deep generative models have emerged as a powerful class of priors for signals in various inverse problems such as compressed sensing, phase retrieval and super-resolution.
Differentiable Programming for Piecewise Polynomial Functions
The paradigm of differentiable programming has considerably enhanced the scope of machine learning via the judicious use of gradient-based optimization.
Adversarially Robust Learning via Entropic Regularization
In this paper we propose a new family of algorithms, ATENT, for training adversarially robust deep neural networks.
Deep Generative Models that Solve PDEs: Distributed Computing for Training Large Data-Free Models
Here we report on a software framework for data parallel distributed deep learning that resolves the twin challenges of training these large SciML models - training in reasonable time as well as distributing the storage requirements.
Hyperparameter Optimization in Neural Networks via Structured Sparse Recovery
In this paper, we study two important problems in the automated design of neural networks -- Hyper-parameter Optimization (HPO), and Neural Architecture Search (NAS) -- through the lens of sparse recovery methods.
ESPN: Extremely Sparse Pruned Networks
Deep neural networks are often highly overparameterized, prohibiting their use in compute-limited systems.
Nano-Material Configuration Design with Deep Surrogate Langevin Dynamics
We consider the problem of optimizing by sampling under multiple black-box constraints in nano-material design.
Benefits of Jointly Training Autoencoders: An Improved Neural Tangent Kernel Analysis
Starting from a randomly initialized autoencoder network, we rigorously prove the linear convergence of gradient descent in two learning regimes, namely: (i) the weakly-trained regime where only the encoder is trained, and (ii) the jointly-trained regime where both the encoder and the decoder are trained.
On Higher-order Moments in Adam
In this paper, we investigate the popular deep learning optimization routine, Adam, from the perspective of statistical moments.
Surrogate-Based Constrained Langevin Sampling With Applications to Optimal Material Configuration Design
We consider the problem of generating configurations that satisfy physical constraints for optimal material nano-pattern design, where multiple (and often conflicting) properties need to be simultaneously satisfied.
Phase Retrieval using Untrained Neural Network Priors
Untrained deep neural networks as image priors have been recently introduced for linear inverse imaging problems such as denoising, super-resolution, inpainting and compressive sensing with promising performance gains over hand-crafted image priors such as sparsity.
Spatiotemporally Constrained Action Space Attacks on Deep Reinforcement Learning Agents
In this work, we first frame the problem as an optimization problem of minimizing the cumulative reward of an RL agent with decoupled constraints as the budget of attack.
Algorithmic Guarantees for Inverse Imaging with Untrained Network Priors
Specifically, we consider the problem of solving linear inverse problems, such as compressive sensing, as well as non-linear problems, such as compressive phase retrieval.
One-Shot Neural Architecture Search via Compressive Sensing
Neural Architecture Search remains a very challenging meta-learning problem.
Encoding Invariances in Deep Generative Models
Reliable training of generative adversarial networks (GANs) typically require massive datasets in order to model complicated distributions.
Reducing The Search Space For Hyperparameter Optimization Using Group Sparsity
We propose a new algorithm for hyperparameter selection in machine learning algorithms.
Semantic Adversarial Attacks: Parametric Transformations That Fool Deep Classifiers
We propose a novel approach to generate such `semantic' adversarial examples by optimizing a particular adversarial loss over the range-space of a parametric conditional generative model.
A Kaczmarz Algorithm for Solving Tree Based Distributed Systems of Equations
We introduce a modified Kaczmarz algorithm for solving systems of linear equations in a distributed environment, i. e. the equations within the system are distributed over multiple nodes within a network.
Alternating Phase Projected Gradient Descent with Generative Priors for Solving Compressive Phase Retrieval
We empirically show that the performance of our method with projected gradient descent is superior to the existing approach for solving phase retrieval under generative priors.
Signal Reconstruction from Modulo Observations
We consider the problem of reconstructing a signal from under-determined modulo observations (or measurements).
Physics-aware Deep Generative Models for Creating Synthetic Microstructures
The first model is a WGAN model that uses a finite number of training images to synthesize new microstructures that weakly satisfy the physical invariances respected by the original data.
Algorithmic Aspects of Inverse Problems Using Generative Models
The traditional approach of hand-crafting priors (such as sparsity) for solving inverse problems is slowly being replaced by the use of richer learned priors (such as those modeled by generative adversarial networks, or GANs).
Learning ReLU Networks via Alternating Minimization
We propose and analyze a new family of algorithms for training neural networks with ReLU activations.
Autoencoders Learn Generative Linear Models
For each of these models, we prove that under suitable choices of hyperparameters, architectures, and initialization, autoencoders learned by gradient descent can successfully recover the parameters of the corresponding model.
On Consensus-Optimality Trade-offs in Collaborative Deep Learning
In distributed machine learning, where agents collaboratively learn from diverse private data sets, there is a fundamental tension between consensus and optimality.
On Learning Sparsely Used Dictionaries from Incomplete Samples
Second, we propose an initialization algorithm that utilizes a small number of extra fully observed samples to produce such a coarse initial estimate.
Solving Linear Inverse Problems Using GAN Priors: An Algorithm with Provable Guarantees
In this work, we advocate the idea of replacing hand-crafted priors, such as sparsity, with a Generative Adversarial Network (GAN) to solve linear inverse problems such as compressive sensing.
Fast Low-Rank Matrix Estimation without the Condition Number
In this paper, we provide a novel algorithmic framework that achieves the best of both worlds: asymptotically as fast as factorization methods, while requiring no dependency on the condition number.
Fast, Sample-Efficient Algorithms for Structured Phase Retrieval
For this problem, we design a recovery algorithm that we call Block CoPRAM that further reduces the sample complexity to O(ks log n).
A Forward-Backward Approach for Visualizing Information Flow in Deep Networks
We introduce a new, systematic framework for visualizing information flow in deep networks.
Provably Accurate Double-Sparse Coding
To our knowledge, our work introduces the first computationally efficient algorithm for double-sparse coding that enjoys rigorous statistical guarantees.
Reconstruction from Periodic Nonlinearities, With Applications to HDR Imaging
We consider the problem of reconstructing signals and images from periodic nonlinearities.
Demixing Structured Superposition Signals from Periodic and Aperiodic Nonlinear Observations
We consider the demixing problem of two (or more) structured high-dimensional vectors from a limited number of nonlinear observations where this nonlinearity is due to either a periodic or an aperiodic function.
Fast Algorithms for Learning Latent Variables in Graphical Models
Existing methods for this problem assume that the precision matrix of the observed variables is the superposition of a sparse and a low-rank component.
Collaborative Deep Learning in Fixed Topology Networks
There is significant recent interest to parallelize deep learning algorithms in order to handle the enormous growth in data and model sizes.
Improved Algorithms for Matrix Recovery from Rank-One Projections
We consider the problem of estimation of a low-rank matrix from a limited number of noisy rank-one projections.
Sample-Efficient Algorithms for Recovering Structured Signals from Magnitude-Only Measurements
For this problem, we design a recovery algorithm Block CoPRAM that further reduces the sample complexity to $O(ks\log n)$.
Iterative Thresholding for Demixing Structured Superpositions in High Dimensions
Specifically, we show that for certain types of structured superposition models, our method provably recovers the components given merely $n = \mathcal{O}(s)$ samples where $s$ denotes the number of nonzero entries in the underlying components.
Stable Recovery Of Sparse Vectors From Random Sinusoidal Feature Maps
Random sinusoidal features are a popular approach for speeding up kernel-based inference in large datasets.
Fast recovery from a union of subspaces
We address the problem of recovering a high-dimensional but structured vector from linear observations in a general setting where the vector can come from an arbitrary union of subspaces.
Fast Algorithms for Demixing Sparse Signals from Nonlinear Observations
We study the problem of demixing a pair of sparse signals from noisy, nonlinear observations of their superposition.
Efficient Upsampling of Natural Images
1) For the edge layer, we use a nonparametric approach by constructing a dictionary of patches from a given image, and synthesize edge regions in a higher-resolution version of the image.
Sparse Signal Recovery Using Markov Random Fields
Compressive Sensing (CS) combines sampling and compression into a single sub-Nyquist linear measurement process for sparse and compressible signals.
Random Projections for Manifold Learning
First, we show that with a small number $M$ of {\em random projections} of sample points in $\reals^N$ belonging to an unknown $K$-dimensional Euclidean manifold, the intrinsic dimension (ID) of the sample set can be estimated to high accuracy.
