Survey of challenges and future directions in applying RL to real-world settings.
problem Challenges in deploying RL in practical settings due to limited interaction and changing environments.
method Analysis of RL system design, implementation, and continual improvement.
result Need for theory and methodology to bridge research and application gap.
Underspecified ML models can behave unpredictably in real-world use.
problem ML models can fail in real-world deployment due to ambiguous predictors.
method Identified underspecification as the cause, showing it affects various ML domains.
result Underspecified models can behave differently in deployment domains.
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.
GRAM enhances deep RL for reliable real-world deployment.
problem Generalizing deep RL across in-distribution and out-of-distribution scenarios.
method Introduces a robust adaptation module and a joint training pipeline.
result GRAM achieves strong generalization performance in simulations and hardware.
Deep Reinforcement Learning (DRL) has become increasingly powerful in recent years, with notable achievements such as Deepmind's AlphaGo. It has been successfully deployed in commercial vehicles like Mobileye's path planning system. However, a vast majority of work on DRL is focused on toy examples in controlled synthe…
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.
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.
Suitability filter detects model performance degradation in real-world deployment.
problem Ensuring model reliability in safety-critical domains without access to ground truth labels.
method Uses suitability signals to evaluate classifier performance on unlabeled user data.
result The suitability filter reliably detects performance deviations due to covariate shift.
Self-supervised policy adapts after deployment without rewards.
problem Generalizing reinforcement learning policies across different environments.
method Uses self-supervision to train policies in new environments without reward signals.
result Significant improvements in generalization across diverse environments.
BREMEN optimizes policies offline with fewer data, achieving efficient deployment.
problem High cost of updating policies in real-world applications.
method Behavior-Regularized Model-ENsemble (BREMEN) algorithm for offline optimization.
result BREMEN achieves impressive deployment efficiency with 5-10 deployments, outperforming standard RL methods.
BoFire optimizes chemistry experiments using Bayesian Optimization.
problem Effective deployment of Bayesian Optimization in the chemical industry.
method Combines Bayesian Optimization with DoE strategies, providing a rich feature-set.
result BoFire enables seamless integration into RESTful APIs for real-world use.
Optimizes electric aircraft deployment for Canadian aviation to reduce emissions.
problem Limited fleet capacity and operational structure hinder electric aircraft transition.
method Multi-period mixed-integer linear programming (MILP) framework.
result Electric aircraft can reduce emissions by over 70% within five years.
This paper improves STL inference reliability under covariate shift.
problem Ensuring correct STL formulas in real-world settings with distribution shift.
method Proposes a conformalized STL inference framework that addresses covariate shift.
result Significantly improves symbolic learning reliability at deployment time.
Picket guards against corrupted data in machine learning models.
problem Data corruption biases models and invalidates predictions.
method PicketNet detects corrupted data using self-supervised deep learning; flags corrupted queries online.
result Picket consistently protects models from corrupted data during training and deployment.
Geometric stability predicts steerability and detects drift in language models.
problem Predicting steerability and detecting drift in language models.
method Supervised and unsupervised geometric stability measures.
result Supervised geometric stability predicts steerability with high accuracy and detects drift earlier.
New method optimizes policies in non-stationary environments.
problem Optimizing policies in non-stationary, context-dependent environments.
method Two-phase approach: offline learning and online adaptation.
result Our method outperforms existing approaches in both synthetic and real-world datasets.
Graph Neural Networks improve financial fraud detection.
problem Complex financial transactions pose challenges in fraud detection.
method Unified framework of GNN methodologies applied to financial fraud detection.
result GNNs excel at capturing complex relational patterns in financial networks.
FLBench automates federated learning benchmarking.
problem Manual dataset partitioning fails to simulate real-world isolated data islands.
method Develops a federated learning benchmark suite with three domains.
result Automates evaluation of federated learning systems and algorithms.
Risk Advisor predicts and mitigates ML deployment failures.
problem Predicting and mitigating test-time failure risks of ML systems.
method Post-hoc meta-learner for estimating failure risks and uncertainties.
result Reliably predicts deployment-time failure risks across various ML models.
The paper tackles fairness and accuracy in ML models under domain shifts.
problem Designing fair and accurate ML models that perform well in unseen domains.
method Theoretical bounds and sufficient conditions for fairness and accuracy transfer under domain generalization.
result A learning algorithm that ensures fair and accurate models even when deployment environments change.
Paper discusses challenges in deploying ML models for structural engineering.
problem Challenges in deploying machine learning models for structural engineering applications.
method Illustrates challenges through two examples, focusing on model overfitting, underspecification, training data representativeness, variable omission bias, and cross-validation.
result Highlights the importance of rigorous model validation techniques.
DEPLOYERS models multi-country economic systems using ABM.
problem Simulate complex multi-country economic systems with detailed data.
method Agent-based modeling framework with multi-threaded simulation.
result Simulates thousands of individuals and firms in a realistic environment.
Survey finds many adversarial machine learning threats are not critical for most entities.
problem Adversarial machine learning threats and their impact on model accuracy.
method Literature review and analysis of real-world occurrences of adversarial attacks.
result Many adversarial machine learning threats do not warrant the cost of robust models.
Framework for applying GPs to real-world data with scalability guidelines.
problem Deployment of Gaussian Processes (GPs) is hindered by computational costs and lack of guidelines.
method Proposed a framework for identifying GP suitability and setting up robust models, formalizing decisions of experienced practitioners.
result More accurate results at test time for glacier elevation change case study.
Machine learning (ML) techniques are enjoying rapidly increasing adoption. However, designing and implementing the systems that support ML models in real-world deployments remains a significant obstacle, in large part due to the radically different development and deployment profile of modern ML methods, and the range …
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.
This thesis tackles NILM challenges with a new dataset and efficient edge deployment techniques.
problem Limited datasets and high computational power for NILM deployment.
method Developed an interoperable data collection framework and introduced model compression techniques.
result Efficient edge deployment of NILM models for global scalability and sustainability.
Safe reinforcement learning framework using optimal transport for robustness.
problem Robustness and safety in deep reinforcement learning with limited data assumptions.
method Optimal transport perturbations to construct worst-case virtual state transitions.
result Significantly improved safety at deployment time compared to standard methods.
Algorithm mitigates performance loss in constrained reinforcement learning with model misspecification.
problem Performance loss in reinforcement learning policies due to model misspecification in constrained control systems.
method Proposes an algorithm to handle constrained model misspecification in continuous control systems.
result Algorithm successfully mitigates performance loss in real-world reinforcement learning tasks.
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…
PRoA assesses deep learning robustness against practical functional perturbations.
problem Inadequate practical robustness verification methods for deep learning systems.
method Probabilistic robustness assessment based on adaptive concentration.
result Statistical guarantees on probabilistic robustness against functional perturbations.
Focuses on monitoring and explaining models in real-world applications.
problem Ensuring high quality machine learning services in production environments.
method Statistical techniques for model performance and data monitoring, explanations of predictions.
result Challenges and solutions for implementing monitoring and explanation in production models.
A novel method quantizes Batch Normalization for QNNs, maintaining accuracy and efficiency.
problem Quantization challenges in Batch Normalization for QNNs.
method Converts BN to fixed-point operation with shared scale, suitable for hardware.
result Maintains same outputs through rigorous analysis and experiments.
Traditional control methods are inadequate in many deployment settings involving control of Cyber-Physical Systems (CPS). In such settings, CPS controllers must operate and respond to unpredictable interactions, conditions, or failure modes. Dealing with such unpredictability requires the use of executive and cognitive…
Infrastructure monitors AI/ML radiology models across multiple sites.
problem Monitoring and improving AI/ML radiology models across multiple sites.
method Interactive radiology reporting, centralized cloud system, post-marketing surveillance.
result Efficient monitoring and iterative development of AI/ML models without radiologist burden.
The discovery of adversarial examples has raised concerns about the practical deployment of deep learning systems. In this paper, we demonstrate that adversarial examples are capable of manipulating deep learning systems across three clinical domains. For each of our representative medical deep learning classifiers, bo…
The workshop focuses on AI principles for structured data.
problem Using AI on structured data for decision-making.
method Addressing principles of privacy, accountability, interpretability, robustness, and reasoning.
result Designing approaches to use structured data for reliable decisions.
A method for quickly determining deployment schedules that meet a given fuel cycle demand is presented here. This algorithm is fast enough to perform in situ within low-fidelity fuel cycle simulators. It uses Gaussian process regression models to predict the production curve as a function of time and the number of depl…
The paper develops methods to reduce deployment risk under dynamic covariate shifts.
problem Reduction of deployment risk under dynamic covariate shifts.
method Time-domain Poincare inequality and Jacobian-velocity theorem to identify and control directional tangent energy.
result Drift-aligned tangent regularization (DTR) reduces risk volatility and directional gain in low-rank drift regimes.
Computer vision performances have been significantly improved in recent years by Convolutional Neural Networks(CNN). Currently, applications using CNN algorithms are deployed mainly on general purpose hardwares, such as CPUs, GPUs or FPGAs. However, power consumption, speed, accuracy, memory footprint, and die size sho…
Joint training improves model accuracy by selectively using privileged information.
problem Two-stage training can lead to model failure with noisy privileged information.
method Joint training of two models to use privileged information selectively.
result Joint training outperforms two-stage baselines on synthetic and real-world tasks.
Robust OPE framework uses human inputs to improve policy evaluation in changing environments.
problem Inaccurate policy evaluations due to shifts in environment properties.
method Adapts OPE methods to shifts on user-inputted covariates, providing more realistic utility estimates.
result Robust OPE framework yields less pessimistic policy evaluations and captures realistic dataset shifts.
New multi-spiral approach improves packaging of thick membranes.
problem Deploying thick membranes on curved surfaces efficiently.
method Multi-spiral folding approach with prismatic folding lines.
result Improved deployment performance of thick membranes on curved surfaces.
POLA adapts learning rates for online time series prediction.
problem Adapting to changing data distributions in dynamic environments.
method Adaptive learning rate regulation for recurrent neural networks.
result POLA outperforms other online prediction methods in real-world datasets.
RL research overhypes potential but lacks deployable solutions.
problem Current RL research overhypes potential and lacks deployable solutions.
method Identifies and critiques current RL research practices.
result Current RL research direction is unlikely to lead to practical, economically viable solutions.
MLDemon monitors ML systems post-deployment, improving reliability with real-time performance estimates and expert labels.
problem Ensuring reliability of machine learning systems post-deployment, especially when user inputs differ from training data.
method Integrates unlabeled and on-demand labeled data to monitor ML model performance in real-time, deciding when to acquire expert labels.
result Outperforms existing approaches in temporal datasets with diverse distribution drifts, providing theoretical optimality for distribution drifts.
New RL algorithms improve control tasks with data reuse.
problem Real-world control requires performance guarantees and data efficiency.
method Generalized Policy Improvement combining on-policy guarantees and sample reuse.
result Extensive experimental analysis shows benefits of new algorithms.
CSA fills a gap in RLVR-trained LLM deployment by providing anytime-valid selective risk control.
problem Deployment of RLVR-trained LLMs in regulated organizations requires a safety certificate for every round without waiting for long-run averages.
method CSA uses a (test statistic, validity guarantee, deployment rule) framework to fill the gap, maintaining a Ville-type e-process per threshold on a Bonferroni grid.
result CSA provides the first anytime-valid selective risk control for RLVR-trained LLMs, matching the long-run average certification rate and satisfying pathwise validity and non-refusing deployment on every cell.