Paper tackles continual reinforcement learning challenges with diversity exploration and adversarial self-correction.
problem Challenges in learning new tasks sequentially due to catastrophic forgetting.
method Develops an end-to-end framework (CDAN) combining unsupervised diversity exploration and adversarial self-correction.
result Final result outperforms baseline by 18.35% in NSD and 0.61 in average reward.
Theoretical analysis shows LLMs can self-correct responses through in-context learning.
problem Understanding how large language models improve through self-correction.
method Theoretical analysis based on simplified alignment task, focusing on softmax attention, multi-head attention, and MLP blocks.
result LLMs can refine responses in an in-context way when given accurate self-examinations as rewards.
SCoreBO improves Bayesian optimization by learning hyperparameters and self-correcting.
problem Efficient hyperparameter tuning for Gaussian process models in Bayesian optimization.
method Introduces SAL and SCoreBO, which prioritize hyperparameter learning and perform simultaneous optimization and learning.
result SCoreBO outperforms state-of-the-art methods on traditional benchmarks and atypical tasks.
Training-free method improves text-to-image generation quality.
problem Error accumulation in simultaneous token updates of masked diffusion models.
method Training-free self-correction framework exploiting inductive biases.
result Significantly improved generation quality on text-to-image tasks.
YOASOVI improves stochastic VI for large models with fast, self-correcting sampling.
problem Efficiently performing stochastic Variational Inference on large Bayesian models.
method YOASOVI uses acceptance sampling to draw only one sample per iteration, improving convergence speed and accuracy.
result YOASOVI converges faster and more accurately than regular Monte Carlo and Quasi-Monte Carlo methods.
Paper stabilizes generative model training with synthetic data.
problem Self-consuming loops in generative model training.
method Introducing an idealized correction function and self-correction functions.
result Self-consuming loops can be exponentially more stable with the right correction.
This work explores test-time scaling strategies for LLMs, improving sample efficiency and expressiveness.
problem Understanding the sample efficiency and expressiveness of test-time scaling strategies for LLMs.
method Established separation and expressiveness results for self-consistency, best-of-n, and self-correction strategies. result Self-correction enables Transformers to simulate online learning over multiple tasks without prior knowledge.
Random feature model shows slow self-correction of generalization gap.
problem Slow deterioration of generalization error in random feature model.
method Examined the dynamic behavior of gradient descent in the model's resonance regime.
result Gradient descent exhibits a self-correction mechanism, reducing generalization gap over time.
Building a large image dataset with high-quality object masks for semantic segmentation is costly and time consuming. In this paper, we introduce a principled semi-supervised framework that only uses a small set of fully supervised images (having semantic segmentation labels and box labels) and a set of images with onl…
In critical care, intensivists are required to continuously monitor high dimensional vital signs and lab measurements to detect and diagnose acute patient conditions. This has always been a challenging task. In this study, we propose a novel self-correcting deep learning prediction approach to address this challenge. W…
Algorithm corrects mistakes in imitation learning tasks.
problem Covariate shift problem in learning from demonstrations.
method Value Iteration with Negative Sampling (VINS) for conservative extrapolation.
result VINS corrects mistakes of the behavioral cloning policy.
Generative model generates images with multiple object classes.
problem Generating images with multiple object classes.
method Conditional Deep Convolutional GAN architecture, stabilized against collapse.
result System generates diverse samples with inter-object relationships.
Machine learning (ML) training algorithms often possess an inherent self-correcting behavior due to their iterative-convergent nature. Recent systems exploit this property to achieve adaptability and efficiency in unreliable computing environments by relaxing the consistency of execution and allowing calculation errors…
In this paper, we obtain the finite-horizon and infinite-horizon ruin probability asymptotics for risk processes with claims of subexponential tails for non-stationary arrival processes that satisfy a large deviation principle. As a result, the arrival process can be dependent, non-stationary and non-renewal. We give t…
SRPO improves AI alignment with human preferences through self-improvement and task-independent optimization.
problem AI models trained with RLHF lack self-correction mechanisms and struggle with task generalization.
method SRPO formulates the preference learning problem as a min-max objective, optimizing a self-improvement policy and a generative policy in an adversarial fashion, making the solution task-independent.
result SRPO outperforms existing methods, achieving 90% AI Win-Rate on XSum and 56% on Arena-Hard prompts after a single revision.
We provide further evidence that markets trend on the medium term (months) and mean-revert on the long term (several years). Our results bolster Black's intuition that prices tend to be off roughly by a factor of 2, and take years to equilibrate. The story behind these results fits well with the existence of two types …
New method reduces costs in text classification with partial labels.
problem Text classification with limited labeled data.
method Proposes a new maximum likelihood estimator with self-correction for ambiguous training examples.
result Our estimators converge faster under certain conditions.
Accurate real-time tracking of influenza outbreaks helps public health officials make timely and meaningful decisions that could save lives. We propose an influenza tracking model, ARGO (AutoRegression with GOogle search data), that uses publicly available online search data. In addition to having a rigorous statistica…
Convolutional Neural Networks have been a subject of great importance over the past decade and great strides have been made in their utility for producing state of the art performance in many computer vision problems. However, the behavior of deep networks is yet to be fully understood and is still an active area of re…
Paper develops a neural network method for censored survival analysis.
problem Distribution-free quantile prediction for censored survival data.
method Develops a novel neural network algorithm for simultaneous quantile optimization.
result The algorithm produces better calibrated quantiles on real datasets.
New method uses surrogate gradients to train efficient spiking networks on neuromorphic hardware.
problem Training high-performing spiking networks on analog neuromorphic hardware is challenging due to device mismatch and lack of efficient algorithms.
method Introduces a general in-the-loop learning framework based on surrogate gradients.
result Learning self-corrects for device mismatch, resulting in competitive spiking network performance.
In this paper we propose a novel index to quantify and measure the flow of information on macro and micro scales. We discuss the implications of this index for knowledge management fields and also as intellectual capital that can thus be utilized by entrepreneurs. We explore different function and human oriented metric…
Study bandit problems under censorship, estimating performance loss.
problem Estimating performance loss in bandit problems with censored feedback.
method Introduced a broad class of censorship models and analyzed their effective dimension.
result Effective dimension naturally leads to results analogous to uncensored settings.
A new method for math reasoning that allows for iterative correction.
problem Standard reasoning models commit to each token and cannot recover from early errors.
method Generative framework with latent thought vectors for iterative self-correction.
result 30 rethinking iterations surpass baselines with 15 times more parameters.
Proposes a new adversarial model to avoid accuracy vs. adversarial accuracy tradeoff.
problem Inherent tradeoff between accuracy and adversarial accuracy in existing adversarial robustness definitions.
method Introduces Voronoi-epsilon adversary that balances perturbation constraints.
result Voronoi-epsilon adversary avoids accuracy vs. adversarial accuracy tradeoff even with large ε. Adversarial consistency depends on the uniqueness of adversarial Bayes classifiers.
problem Consistency of adversarial surrogate losses is not guaranteed.
method Connected consistency of adversarial surrogate losses to the uniqueness of adversarial Bayes classifiers.
result A convex surrogate loss is statistically consistent for adversarial learning if and only if the adversarial Bayes classifier is unique.
New method makes neural networks more resilient to location-optimized adversarial patches.
problem Neural networks' vulnerability to adversarial patches that are visible but still effective.
method Developed a practical approach to optimize patch locations and applied adversarial training.
result Significantly improved robustness against adversarial patches on CIFAR10 and GTSRB.
Adversarial training achieves optimal test error for shallow networks.
problem Achieving optimal adversarial test error for general data distributions.
method Applying new Rademacher complexity bounds and properties of optimal adversarial predictors.
result Adversarial training can achieve optimal adversarial test error for general data distributions.
SPAT improves adversarial robustness by preserving semantics in adversarial training.
problem Adversarial examples often have different semantics than original data, introducing unintended biases.
method Semantics-preserving adversarial training (SPAT) that encourages pixel perturbation shared among all classes.
result SPAT improves adversarial robustness and achieves state-of-the-art results in CIFAR-10 and CIFAR-100.
New approach improves adversarial robustness without sacrificing natural generalization.
problem Balancing adversarial robustness and natural generalization in machine learning.
method Friendly adversarial training (FAT) using early-stopped PGD to find least adversarial data.
result Early-stopped PGD achieves adversarial robustness without compromising natural generalization.
Adversarial training adds dynamic perturbations to neural networks for robustness.
problem Accuracy trade-off and lack of diversity in adversarial examples.
method Dynamic adversarial perturbations in the parameter space of neural networks, updating perturbation biases during training.
result Adversarial training with negligible cost and reduced accuracy trade-off.
Simple regularization methods mimic adversarial training's robustness.
problem Expensive adversarial training for robustness.
method Label smoothing and logit squeezing.
result Achieves strong adversarial robustness without adversarial examples.
By injecting adversarial examples into training data, adversarial training is promising for improving the robustness of deep learning models. However, most existing adversarial training approaches are based on a specific type of adversarial attack. It may not provide sufficiently representative samples from the adversa…
Paper analyzes adversarial risk using optimal transport.
problem Poor performance of machine learning on adversarial data.
method Optimal transport perspective, optimal transport plans (couplings), convexity, smoothness assumptions.
result Fundamental limits on adversarial risk calculated for various datasets.
Paper bounds convergence rate of adversarial surrogate risk.
problem Vulnerability of binary classification models to adversarial attacks.
method Characterizes conditions for adversarial consistency and provides surrogate risk bounds.
result Surrogate risk bounds quantify the rate of convergence of adversarial classification risk.
Study on robustness in linear regression models, focusing on adversarial perturbations.
problem Understanding and improving robustness in linear regression models to adversarial perturbations.
method Developed a two-stage adversarial learning framework that incorporates model structure information.
result Proved the consistency and developed the Bahadur representation of the adversarially robust estimator.
Localized adversarial training improves image classifiers' robustness.
problem State-of-the-art image classifiers fail on carefully manipulated adversarial images.
method Developed a localized adversarial attack and used it to train a robust classifier.
result Localized adversarial training increases robustness against adversarial inputs.
Ideally, what confuses neural network should be confusing to humans. However, recent experiments have shown that small, imperceptible perturbations can change the network prediction. To address this gap in perception, we propose a novel approach for learning robust classifier. Our main idea is: adversarial examples for…
A Bayesian framework models adversarial uncertainty for robust machine learning.
problem Vulnerability of machine learning models to adversarial attacks.
method Formal Bayesian framework that models adversarial uncertainty through a stochastic channel, articulating probabilistic assumptions.
result Explicitly modeling adversarial uncertainty leads to improved robustification strategies.
Adversarial training improves audio classifier robustness.
problem Improving robustness of audio classifiers against adversarial attacks.
method Adversarial training of deep neural networks on audio data.
result Adversarially trained models can resist higher perturbations and fooling rates.
Neural networks are vulnerable to adversarial attacks -- small visually imperceptible crafted noise which when added to the input drastically changes the output. The most effective method of defending against these adversarial attacks is to use the methodology of adversarial training. We analyze the adversarially train…
Paper tackles robustness in adversarial noise with a meta-optimizer.
problem Sensitivity to adversarial noise hinders machine learning deployment.
method Meta-optimizer learns to robustly optimize models using adversarial examples.
result Meta-optimizer transfers adversarial knowledge to new models without generating new examples.
New adversarial examples with structured distortion sets improve robustness and perceptibility.
problem Improving adversarial robustness and perceptibility of images.
method Exploring and constraining adversarial search with trace-norms and other norms.
result Structured adversarial perturbations allow larger distortions and control over generation.
Survey of adversarial examples in visual machine learning models.
problem Vulnerability of machine learning models to adversarial examples.
method Exploration of various adversarial attack and defense methods.
result Understanding of adversarial attack and defense mechanisms.
Robust RL with learned optimal adversary improves agent performance under adversarial state observations.
problem Ensuring reinforcement learning agents' robustness against adversarial perturbations of state observations.
method Proposed a framework of alternating training with learned adversaries (ATLA) to find optimal adversarial policies and enhance agent robustness.
result ATLA achieves state-of-the-art performance under strong adversaries in continuous control environments.
Interpolated adversarial training improves robustness without sacrificing accuracy.
problem Improving adversarial robustness without degrading generalization performance.
method Interpolated adversarial training using interpolation-based methods.
result Retains adversarial robustness while achieving a standard test error of 6.45%.
Enhances single-step adversarial training to defend against iterative adversarial examples.
problem Defending against iterative adversarial examples in neural networks.
method Identified and leveraged empirical properties of Iter-Adv to improve Single-Adv.
result Enhanced Single-Adv to defend against iterative adversarial examples with improved accuracy and reduced training cost.
In recent years, deep neural networks have demonstrated outstanding performance in many machine learning tasks. However, researchers have discovered that these state-of-the-art models are vulnerable to adversarial examples: legitimate examples added by small perturbations which are unnoticeable to human eyes. Adversari…