Deployment-complete benchmarking assesses if evidence leads to consistent deployment actions.
problem Lack of clear evidence leading to consistent deployment actions.
method Introduces deployment-complete benchmarking to test if benchmark evidence determines deployment actions.
result Benchmark evidence must be complete for a claim to lead to a consistent deployment action.
A framework to quantify deployment risk in ML systems, especially for rare states.
problem Under-supported rare states in ML models lead to unreliable performance in unseen data.
method Blind-Spot Mass (B_n(tau)) using Good-Turing unseen-species estimation.
result Identifies and quantifies the risk of under-supported states in ML models.
New active learning method uses combinatorial coverage to improve data transfer and reduce bias.
problem Inability to transfer sampled data to new models and sampling bias issues.
method Data-centric active learning methods utilizing combinatorial coverage.
result Sampling data with coverage leads to better data transfer and competitive sampling bias.
CPATTA uses conformal prediction for efficient test-time adaptation.
problem Low data selection efficiency in existing ATTA methods.
method Conformal Prediction, online weight-update algorithm, domain-shift detector, staged update scheme.
result CPATTA consistently outperforms state-of-the-art methods by 5% in accuracy.
New approach for adaptive conformal inference using Blackwell's theory.
problem Non-exchangeable environments in sequential conformal inference.
method Reinterpretation of ACI as a game, construction of coverage and efficiency objectives, approachability strategy.
result Algorithm achieves strong theoretical guarantees and practical insights.
Survey of reinforcement learning guarantees with data constraints.
problem Guaranteeing near-optimal policies with limited data in reinforcement learning.
method Coverage-Structure-Objective (CSO) framework to decompose sample complexity results.
result Progress on PAC guarantees for reinforcement learning, covering various models and settings.
DS-CP improves reliability of uncertainty quantification for large language models under domain shift.
problem Overconfident and factually incorrect outputs (hallucinations) from large language models.
method Adapts conformal prediction to large language models under domain shift by reweighting calibration samples.
result DS-CP delivers more reliable coverage than standard conformal prediction, especially under substantial distribution shifts.
CF-GNN provides reliable uncertainty estimates for graph data.
problem Lack of rigorous uncertainty estimates in GNNs.
method Conformal prediction extended to graph-based models (CF-GNN).
result CF-GNN produces prediction sets with provable coverage probability.
New method provides formal uncertainty guarantees for image classifiers.
problem Uncertainty quantification for image classifiers without formal guarantees.
method Adapts conformal prediction to give stable, formal coverage guarantees.
result Method outperforms existing approaches in coverage and set size.
Conformal prediction fails under severe feature turnover in COVID-19 supply chain tasks.
problem Dealing with distribution shift in conformal prediction models.
method Using COVID-19 as a natural experiment across 8 supply chain tasks, analyzing SHAP explanations.
result Coverage drops vary widely (0% to 86.7%) and correlate with single-feature dependence.
CONTINA provides adaptive confidence intervals for traffic demand prediction.
problem Uncertainty in future traffic demand predictions and the need for valid confidence intervals.
method Adaptive confidence interval method that adjusts based on deployment errors.
result Valid confidence intervals with shorter lengths and theoretical coverage guarantee.
SConU improves uncertainty prediction for large models by testing for outliers and reducing miscoverage.
problem Real-world deployment of large language models requires reliable guarantees of task-specific metrics.
method SConU implements significance tests to identify and exclude outliers that violate exchangeability assumptions.
result SConU reduces miscoverage rates and enhances prediction efficiency in high-stakes tasks.
Study bandwidth-limited training and inference of language models.
problem Training and inference of language models on scattered data with limited bandwidth.
method Analyzed two protocols: Federated Probe-Logit Distillation (FPLD) for training and Federated Conformal RAG (FC-RAG) for inference.
result Explicit high-probability KL-consistency rate and distribution-free marginal-coverage bound for Federated Conformal RAG.
Paper proposes a graph model for optimal AP deployment in indoor optical wireless networks.
problem Challenges in deploying optical wireless networks due to LoS requirement and limited range.
method Graph modeling approach to identify minimum number of APs and their optimal locations.
result Optimal deployment of APs ensures connectivity and minimizes interference in indoor environments.
New algorithm provides robust uncertainty quantification without parameter tuning.
problem Real-world machine learning predictors need reliable uncertainty quantification.
method Parameter-free, group-conditional online prediction algorithm.
result Achieves best group-conditional coverage guarantees.
Study online RL with mismatched dynamics, achieving sublinear regret.
problem Exploration challenges in online RL with mismatched training and deployment dynamics.
method Introduce supremal visitation ratio, propose efficient algorithm with f-divergence. result Achieves sublinear regret in online RMDPs with optimal dependence on supremal visitation ratio and interaction episodes.
This thesis tackles data imperfections in ML, proposing methods to prevent discrimination and spurious feature learning.
problem Data imperfections in ML training data, leading to deployment failures.
method Fair representation learning, spurious feature detection, and data augmentation for reinforcement learning.
result Methods to make ML models more robust to data imperfections.
The paper certifies AI reliability via sampling and calibration, providing exact guarantees.
problem Ensuring trust in black-box AI systems' outputs.
method Self-consistency sampling and conformal calibration.
result Reliability levels derived from these methods offer finite-sample guarantees.
Study designs logging policies to minimize off-policy evaluation error.
problem Minimizing OPE error with logging policies for target policies.
method Characterizes reward-coverage tradeoff, proposes a unifying framework, derives optimal policies.
result Provides actionable guidance for firms choosing recommendation systems.
CMCO provides robust uncertainty estimates for neural operators without retraining.
problem Uncertainty quantification in deep learning for real-time virtual sensing.
method Unified Monte Carlo dropout and split conformal prediction in DeepONet.
result Near-nominal empirical coverage in diverse applications.
EPIC quantifies reward differences without policy optimization.
problem Distinguishing reward function quality from policy optimization issues.
method EPIC distance to compare reward functions directly.
result EPIC bounds policy training success and regret.
CJE calibrates cheap LLM judges against an oracle, achieving high accuracy at a fraction of the cost.
problem Inexpensive LLM judges can produce biased rankings, leading to unreliable outcomes.
method CJE uses a small oracle to calibrate cheap scores, then evaluates at scale with valid uncertainty.
result CJE achieves 99% pairwise ranking accuracy at 14x lower cost compared to a 16x oracle/judge cost ratio.
Improves off-policy evaluation with imperfect annotations.
problem Limited dataset coverage for evaluating new policies.
method Doubly robust estimators combining IS and DM, incorporating counterfactual annotations.
result Using annotations within the DM component yields the most desirable theoretical results.
A geometric theory explains loss functions for robust representation learning.
problem Treats robustness, domain adaptation, and sensor drift as separate literatures.
method Estimates covariance Sigma_task and uses it to pin Jacobian penalties.
result Proves optimality and necessity of range coverage for penalty matrices.
CSP improves time-series forecasting without training, outperforming DeepNPTS in speed and accuracy.
problem Improving probabilistic time-series forecasting without training.
method Mixing empirical and residual draws around a seasonal naive forecast.
result CSP significantly outperforms DeepNPTS on CRPS, normalized mean quantile loss, and coverage metrics.
New method improves conformal prediction for machine learning models.
problem Improving the efficiency and informativeness of conformal prediction models.
method Introduces Penalized Inverse Probability (PIP) and Regularized PIP (RePIP) nonconformity score functions.
result PIP-based conformal classifiers strike a good balance between informativeness and efficiency.
New method certifies risks of LLM outputs, improving accuracy and reliability.
problem Uncertain and incorrect outputs from large language models.
method Information-lift certificates using PAC-Bayes bounds and skeleton design.
result Achieves 77.0% coverage at 2% risk, outperforming baselines.
New method uses label-weighted conformal prediction for macro-coverage guarantees in classification.
problem Finding a balance between class-conditional and marginal coverage in long-tailed datasets.
method Label-weighted conformal prediction for macro-coverage guarantees.
result Validated prediction sets with macro-coverage guarantees on large-scale image datasets.
A new method improves efficiency of conformal prediction for ensemble models.
problem Efficiently estimating uncertainty for ensemble models without distributional assumptions.
method Proposes a multivariate score function to merge prediction regions of individual models, reducing conservatism.
result Demonstrates more efficient prediction regions compared to existing methods.
Paper establishes lower bounds and optimal algorithms for deployment-efficient RL.
problem Deployment efficiency in reinforcement learning.
method Optimization with constraints, lower bounds, algorithms.
result Established optimal algorithms for deployment-efficient RL.
DUPLE tackles cross-deployment recognition in fiber-optic perimeter security with meta-learning.
problem Cross-deployment recognition challenges in fiber-optic perimeter security due to label scarcity and distribution shifts.
method DUPLE employs statistically guided meta-learning to enhance recognition robustness across unseen deployments.
result DUPLE consistently outperforms traditional and meta-learning baselines in cross-deployment DFOS benchmarks.
Algorithm balances learning and coverage for multi-robots over unknown fields.
problem Balancing learning and coverage for multi-robots over unknown, nonuniform sensory fields.
method DSLC algorithm that schedules learning and coverage epochs, using Gaussian Process modeling and coverage regret analysis.
result Upper bound on expected cumulative coverage regret provided for DSLC.
Benchmarking 18 deep neural network selective classification models on various datasets.
problem Designing a model that can abstain from making predictions when there's high risk of error.
method Evaluation of 18 selective classification models using multiple criteria.
result No single model is the clear winner; best model depends on objectives.
Improved conformal prediction for better conditional coverage of classifier predictions.
problem Achieving exact conditional coverage in finite samples for prediction sets.
method Developed a variant of conformal prediction targeting coverage conditional on confidence and trust score.
result Empirically improved conditional coverage properties compared to standard conformal prediction.
AM-PPI uses multiple predictors to reduce label cost in healthcare AI.
problem Reduces label cost in post-deployment monitoring of healthcare AI.
method Combines model predictions with a small labeled sample, routing each instance to a cost-appropriate subset of predictors.
result Produces narrower confidence intervals than single-predictor methods.
Adaptive coverage policies improve conformal prediction accuracy.
problem Fixed coverage levels in traditional conformal prediction lead to uninformative predictions.
method Optimizes adaptive coverage policy using a neural network trained on leave-one-out calibration.
result Adaptive coverage policies produce more informative and flexible prediction sets.
Machine learning (ML), especially deep learning is made possible by the availability of big data, enormous compute power and, often overlooked, development tools or frameworks. As the algorithms become mature and efficient, more and more ML inference is moving out of datacenters/cloud and deployed on edge devices. This…
Method predicts biomarker trajectories with uncertainty bands for Alzheimer's disease.
problem Uncertainty in biomarker predictions poses risks in clinical deployment.
method Conformal prediction for randomly-timed biomarker trajectories.
result Conformal bands achieve desired coverage and are tighter than baseline.
New metrics improve understanding of predictive system reliability.
problem Evaluating conditional coverage of predictive systems.
method Casting conditional coverage estimation as a classification problem, using excess risk of the target coverage (ERT) metrics.
result Modern classifiers provide higher statistical power for estimating conditional coverage.
Kandinsky conformal prediction expands conditional coverage guarantees.
problem Disparities in coverage guarantees across different subpopulations.
method Flexible handling of overlapping and fractional group memberships.
result Minimax-optimal high-probability conditional coverage bound.
ST-BCP narrows the coverage gap in BCP by transforming nonconformity scores.
problem The looseness in BCP's coverage guarantee due to Markov's inequality.
method Introduces a data-dependent transformation of nonconformity scores.
result Reduces the average coverage gap from 4.20% to 1.12% on benchmarks.
Improves conditional coverage of regression models using conformal prediction.
problem Lack of conditional coverage guarantees in conformal prediction methods.
method Proposes a novel algorithm to train a regression function to improve conditional coverage after split conformal prediction.
result Establishes an upper bound for miscoverage gap and proposes an end-to-end algorithm to control it.
Small satellite constellations provide daily global coverage of the earth's landmass, but image enrichment relies on automating key tasks like change detection or feature searches. For example, to extract text annotations from raw pixels requires two dependent machine learning models, one to analyze the overhead image …
This study uses AI to analyze financial market coverage from YouTube videos.
problem Challenges in analyzing a large number of financial market videos.
method Used Whisper model to generate text from videos, applied natural language processing.
result Highlights dynamics of financial market coverage and identifies trending topics.
Framework for responsible LLM deployment with human involvement and decentralized technologies.
problem Challenges in deploying LLMs for high-stakes decisions, including data security and accountability.
method Interactive human involvement through multiple iterations, decentralized technologies, and automated auditing.
result Enhanced security and accountability in LLM deployment for financial decisions.
Conformal methods create prediction bands that control average coverage under no assumptions besides i.i.d. data. Besides average coverage, one might also desire to control conditional coverage, that is, coverage for every new testing point. However, without strong assumptions, conditional coverage is unachievable. Giv…
TQA improves prediction intervals for time series data by adjusting quantiles for both cross-sectional and longitudinal coverage.
problem Constructing reliable prediction intervals for cross-sectional time series data.
method Temporal Quantile Adjustment (TQA) method that adjusts the quantile in Conformal Prediction to account for both cross-sectional and longitudinal coverage.
result TQA improves longitudinal coverage while preserving cross-sectional coverage, as validated through extensive experimentation.
This paper evaluates conformal prediction for aerial image classification in challenging environments.
problem Challenging aerial image classification in data-scarce, unconstrained environments.
method Conformal prediction applied to pretrained models (MobileNet, DenseNet, ResNet) with limited labeled data.
result Conformal prediction can provide valuable uncertainty estimates even with small labeled samples.