Search Results for author:
Found 35 papers, 21 papers with code
Adversarial Robustness Against the Union of Multiple Threat Models
Owing to the susceptibility of deep learning systems to adversarial attacks, there has been a great deal of work in developing (both empirically and certifiably) robust classifiers.
Understanding Optimization in Deep Learning with Central Flows
A key difficulty is that much of an optimizer's behavior is implicitly determined by complex oscillatory dynamics, referred to as the "edge of stability."
AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents
The robustness of LLMs to jailbreak attacks, where users design prompts to circumvent safety measures and misuse model capabilities, has been studied primarily for LLMs acting as simple chatbots.
From Variance to Veracity: Unbundling and Mitigating Gradient Variance in Differentiable Bundle Adjustment Layers
We then propose a simple, yet effective solution to reduce the gradient variance by using the weights predicted by the network in the inner optimization loop to weight the correspondence objective in the training problem.
Improving Alignment and Robustness with Circuit Breakers
Existing techniques aimed at improving alignment, such as refusal training, are often bypassed.
Neural Network Verification with Branch-and-Bound for General Nonlinearities
GenBaB is part of the latest $\alpha,\!\beta$-CROWN, the winner of the 4th and the 5th International Verification of Neural Networks Competition (VNN-COMP 2023 and 2024).
Why is SAM Robust to Label Noise?
We infer that SAM's effect in deeper networks is instead explained entirely by the effect SAM has on the network Jacobian.
Forcing Diffuse Distributions out of Language Models
Despite being trained specifically to follow user instructions, today's instructiontuned language models perform poorly when instructed to produce random outputs.
Predicting the Performance of Foundation Models via Agreement-on-the-Line
Recently, it was shown that ensembles of neural networks observe the phenomena "agreement-on-the-line", which can be leveraged to reliably predict OOD performance without labels.
DART: Implicit Doppler Tomography for Radar Novel View Synthesis
Simulation is an invaluable tool for radio-frequency system designers that enables rapid prototyping of various algorithms for imaging, target detection, classification, and tracking.
Generative Posterior Networks for Approximately Bayesian Epistemic Uncertainty Estimation
In many real-world problems, there is a limited set of training data, but an abundance of unlabeled data.
Understanding Augmentation-based Self-Supervised Representation Learning via RKHS Approximation and Regression
Recent work has built the connection between self-supervised learning and the approximation of the top eigenspace of a graph Laplacian operator, suggesting that learning a linear probe atop such representation can be connected to RKHS regression.
Importance of equivariant and invariant symmetries for fluid flow modeling
Graph neural networks (GNNs) have shown promise in learning unstructured mesh-based simulations of physical systems, including fluid dynamics.
Low-Variance Gradient Estimation in Unrolled Computation Graphs with ES-Single
We propose an evolution strategies-based algorithm for estimating gradients in unrolled computation graphs, called ES-Single.
Provably Bounding Neural Network Preimages
Most work on the formal verification of neural networks has focused on bounding the set of outputs that correspond to a given set of inputs (for example, bounded perturbations of a nominal input).
Learning Options via Compression
Identifying statistical regularities in solutions to some tasks in multi-task reinforcement learning can accelerate the learning of new tasks.
Finetune like you pretrain: Improved finetuning of zero-shot vision models
In total, these benchmarks establish contrastive finetuning as a simple, intuitive, and state-of-the-art approach for supervised finetuning of image-text models like CLIP.
Path Independent Equilibrium Models Can Better Exploit Test-Time Computation
Designing networks capable of attaining better performance with an increased inference budget is important to facilitate generalization to harder problem instances.
Improving Adversarial Robustness via Joint Classification and Multiple Explicit Detection Classes
This work concerns the development of deep networks that are certifiably robust to adversarial attacks.
Deep Equilibrium Approaches to Diffusion Models
In this paper, we look at diffusion models through a different perspective, that of a (deep) equilibrium (DEQ) fixed point model.
Efficiently Computing Local Lipschitz Constants of Neural Networks via Bound Propagation
In this paper, we develop an efficient framework for computing the $\ell_\infty$ local Lipschitz constant of a neural network by tightly upper bounding the norm of Clarke Jacobian via linear bound propagation.
Unified Fully and Timestamp Supervised Temporal Action Segmentation via Sequence to Sequence Translation
This paper introduces a unified framework for video action segmentation via sequence to sequence (seq2seq) translation in a fully and timestamp supervised setup.
Ranked #5 on
Action Segmentation
on Assembly101
Test-Time Adaptation via Conjugate Pseudo-labels
In this paper, we start by presenting a surprising phenomenon: if we attempt to meta-learn the best possible TTA loss over a wide class of functions, then we recover a function that is remarkably similar to (a temperature-scaled version of) the softmax-entropy employed by TENT.
Agreement-on-the-Line: Predicting the Performance of Neural Networks under Distribution Shift
In this paper, we show a similar but surprising phenomenon also holds for the agreement between pairs of neural network classifiers: whenever accuracy-on-the-line holds, we observe that the OOD agreement between the predictions of any two pairs of neural networks (with potentially different architectures) also observes a strong linear correlation with their ID agreement.
Defending Multimodal Fusion Models against Single-Source Adversaries
Beyond achieving high performance across many vision tasks, multimodal models are expected to be robust to single-source faults due to the availability of redundant information between modalities.
Understanding Why Generalized Reweighting Does Not Improve Over ERM
Together, our results show that a broad category of what we term GRW approaches are not able to achieve distributionally robust generalization.
Boosted CVaR Classification
To learn such randomized classifiers, we propose the Boosted CVaR Classification framework which is motivated by a direct relationship between CVaR and a classical boosting algorithm called LPBoost.
Exploring Classic and Neural Lexical Translation Models for Information Retrieval: Interpretability, Effectiveness, and Efficiency Benefits
We study the utility of the lexical translation model (IBM Model 1) for English text retrieval, in particular, its neural variants that are trained end-to-end.
You Only Query Once: Effective Black Box Adversarial Attacks with Minimal Repeated Queries
In this work, we instead show that it is possible to craft (universal) adversarial perturbations in the black-box setting by querying a sequence of different images only once.
A community-powered search of machine learning strategy space to find NMR property prediction models
The rise of machine learning (ML) has created an explosion in the potential strategies for using data to make scientific predictions.
Ranked #1 on
NMR J-coupling
on QM9
A FRAMEWORK FOR ROBUSTNESS CERTIFICATION OF SMOOTHED CLASSIFIERS USING F-DIVERGENCES
Formal verification techniques that compute provable guarantees on properties of machine learning models, like robustness to norm-bounded adversarial perturbations, have yielded impressive results.
Provably robust deep generative models
In this paper, we propose a method for training provably robust generative models, specifically a provably robust version of the variational auto-encoder (VAE).
Differentiable Convex Optimization Layers
In this paper, we propose an approach to differentiating through disciplined convex programs, a subclass of convex optimization problems used by domain-specific languages (DSLs) for convex optimization.
On Physical Adversarial Patches for Object Detection
In this paper, we demonstrate a physical adversarial patch attack against object detectors, notably the YOLOv3 detector.
SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver
We demonstrate that by integrating this solver into end-to-end learning systems, we can learn the logical structure of challenging problems in a minimally supervised fashion.
Ranked #1 on
Game of Sudoku
on Sudoku 9x9
