RealCause provides a realistic benchmark for causal inference.
problem Lack of a reliable benchmark for comparing causal effect estimators.
method Flexible generative models to create a benchmark that is both ground-truth and realistic.
result Evaluation of over 1500 causal estimators provides evidence for choosing hyperparameters using predictive metrics.
State-of-the-art machine learning methods exhibit limited compositional generalization. At the same time, there is a lack of realistic benchmarks that comprehensively measure this ability, which makes it challenging to find and evaluate improvements. We introduce a novel method to systematically construct such benchmar…
Unified benchmarks assess data poisoning and backdoor attacks.
problem Unclear danger and effectiveness of data poisoning methods.
method Developed standardized benchmarks for data poisoning and backdoor attacks.
result Existing methods may not generalize to realistic settings.
CausalTime generates realistic time-series for TSCD evaluation.
problem Lack of realistic synthetic datasets for TSCD performance evaluation.
method Harnessing deep neural networks and normalizing flow for dynamics, extracting causal graphs, and deriving ground truth causal graphs.
result Generated datasets accurately reflect real data and ground truth causal graphs.
Benchmark proposes to assess molecule docking efficiency.
problem Lack of realistic benchmarks for measuring progress in drug design.
method Proposes a docking-based benchmark using SMINA software.
result Graph-based generative models fail to generate high-scoring molecules.
SKADA-bench evaluates unsupervised DA methods across diverse modalities.
problem Evaluating unsupervised DA methods on diverse modalities with realistic validation.
method Nested cross-validation and unsupervised model selection scores.
result Highlights the importance of realistic validation and provides practical guidance.
OGB provides diverse graph datasets for robust ML research.
problem Challenges in scalable and robust graph machine learning.
method Unified evaluation protocol, diverse datasets, and automated pipeline.
result Significant scalability and generalization challenges identified.
Because the choice and tuning of the optimizer affects the speed, and ultimately the performance of deep learning, there is significant past and recent research in this area. Yet, perhaps surprisingly, there is no generally agreed-upon protocol for the quantitative and reproducible evaluation of optimization strategies…
Paper introduces benchmark-neutral pricing for long-term contracts.
problem High prices of long-term contracts under risk-neutral pricing.
method Uses growth optimal portfolio as numeraire and new pricing measure.
result Identifies minimal possible prices for contingent claims.
Olympus benchmarks optimization algorithms for noisy experiments.
problem Benchmarking optimization algorithms on realistic experimental scenarios is challenging.
method Introduces Olympus, a software package for benchmarking optimization algorithms on synthetic experiments.
result Mitigates barriers in benchmarking optimization algorithms on realistic experimental scenarios.
SpaCE tackles spatial confounding in scientific studies.
problem Spatial confounding influences treatment and outcome, leading to spurious associations.
method Introduces SpaCE toolkit for benchmark datasets and tools to evaluate causal inference methods.
result Facilitates automated evaluation of machine learning and causal inference models.
New HPO frameworks needed for continual learning.
problem No standard HPO for continual learning.
method Comparative study of HPO frameworks.
result No HPO framework consistently outperforms others.
FinSurvival provides a large-scale financial survival modeling benchmark.
problem Lack of large-scale, realistic, and freely available datasets for benchmarking AI survival models.
method Derived 16 survival modeling tasks from cryptocurrency lending data using an automated pipeline.
result Demonstrated that existing AI survival models are not well-suited for these challenging tasks.
In this article, we describe the algorithms for causal structure learning from time series data that won the Causality 4 Climate competition at the Conference on Neural Information Processing Systems 2019 (NeurIPS). We examine how our combination of established ideas achieves competitive performance on semi-realistic a…
BREEDS benchmarks assess model robustness to subpopulation shifts.
problem Measuring model robustness to novel subpopulation shifts.
method Controlled synthesis of realistic distribution shifts using class structure.
result Validated model sensitivity and effectiveness of robustness interventions.
WILD-SCAV benchmarks AI in complex 3D FPS environments.
problem Lack of complexity and diversity in RL environments.
method Developed a 3D open-world FPS game environment.
result Demonstrates effectiveness in benchmarking RL algorithms.
Paper introduces OARF benchmark suite for federated learning systems.
problem Limited diversity in federated learning benchmarks.
method Characterizes OARF benchmark suite with diverse data and applications.
result Federated learning can effectively increase end-to-end throughput.
Due to the high computational demands executing a rigorous comparison between hyperparameter optimization (HPO) methods is often cumbersome. The goal of this paper is to facilitate a better empirical evaluation of HPO methods by providing benchmarks that are cheap to evaluate, but still represent realistic use cases. W…
Framework generates multimodal datasets with known MI for benchmarking.
problem Benchmarking mutual information estimators and SSL techniques.
method Flow-based generative model and structured causal framework.
result Regression performance improves with increasing MI between modalities.
Generates synthetic data for benchmarking unsupervised outlier detection.
problem Difficulty in benchmarking unsupervised outlier detection due to rare and varied outliers in real data.
method Proposes a generic process to generate synthetic data with insightful characteristics.
result Demonstrates practicality of the generic process through a benchmark with state-of-the-art detection methods.
Synthesizing high-quality images from text descriptions is a challenging problem in computer vision and has many practical applications. Samples generated by existing text-to-image approaches can roughly reflect the meaning of the given descriptions, but they fail to contain necessary details and vivid object parts. In…
Benchmark evaluates financial misinformation detection models, revealing weaknesses without external context.
problem Detecting financial misinformation without external references.
method RFC Bench at paragraph level, two tasks: reference-free detection and comparison-based diagnosis.
result Performance improves with comparative context, revealing model weaknesses in reference-free settings.
Modern federated networks, such as those comprised of wearable devices, mobile phones, or autonomous vehicles, generate massive amounts of data each day. This wealth of data can help to learn models that can improve the user experience on each device. However, the scale and heterogeneity of federated data presents new …
CausalMan simulates complex causality for fair benchmarking.
problem Lack of realistic causal models with known data generating processes.
method Developed a physics-based simulator, CausalMan, for large-scale causality.
result Demonstrated inadequacy of state-of-the-art approaches and analyzed their performance.
We present Simitate --- a hybrid benchmarking suite targeting the evaluation of approaches for imitation learning. A dataset containing 1938 sequences where humans perform daily activities in a realistic environment is presented. The dataset is strongly coupled with an integration into a simulator. RGB and depth stream…
SBSS uses similarity to split data for better classifier training.
problem Training better classifiers with realistic performance estimation.
method SBSS uses both input and output space information to split data using similarity functions.
result SBSS outperformed ordinary stratified 10-fold cross-validation in 75% of scenarios.
The optimization of algorithm (hyper-)parameters is crucial for achieving peak performance across a wide range of domains, ranging from deep neural networks to solvers for hard combinatorial problems. The resulting algorithm configuration (AC) problem has attracted much attention from the machine learning community. Ho…
Study benchmarks embedding-based entity alignment methods for KGs.
problem Align entities across different KGs using embeddings.
method Survey and categorize 23 embedding-based methods, propose new KG sampling algorithm, develop open-source library.
result Understood strengths and limitations of embedding-based methods.
Open dataset and pipeline for realistic OPE research.
problem Lack of realistic and reproducible OPE experimental studies.
method Public logged bandit dataset and Python software.
result Enables experimental comparisons of OPE estimators.
SurvHTE-Bench benchmarks HTE estimation in survival analysis with diverse datasets.
problem Challenges in estimating HTEs from right-censored survival data.
method Modular synthetic datasets, semi-synthetic datasets, and real-world datasets.
result First rigorous comparison of survival HTE methods under diverse conditions.
New benchmarks focus on LLM risk in finance, not just accuracy.
problem Standard benchmarks ignore LLM financial risks, leading to unsafe deployment.
method Three-level agenda: model, workflow, and system stress-testing.
result Hidden weaknesses in LLMs are missed by traditional benchmarks.
Despite the recent progress in hyperparameter optimization (HPO), available benchmarks that resemble real-world scenarios consist of a few and very large problem instances that are expensive to solve. This blocks researchers and practitioners not only from systematically running large-scale comparisons that are needed …
Paper presents a new dataset for testing causal discovery methods in industrial systems.
problem Lack of real-world datasets for evaluating causal discovery methods on time series data.
method Develops a dataset from an industrial system and its known causal graph.
result Provides a benchmark for evaluating causal discovery methods in complex systems.
Paper proposes a new dataset for group anomaly detection in physics.
problem Group anomaly detection in physics.
method Development of a synthetic benchmark dataset and comparison of anomaly detection techniques.
result Demonstrated performance of anomaly detection techniques on the new dataset.
In this paper we propose a novel application of Gaussian processes (GPs) to financial asset allocation. Our approach is deeply rooted in Stochastic Portfolio Theory (SPT), a stochastic analysis framework introduced by Robert Fernholz that aims at flexibly analysing the performance of certain investment strategies in st…
Bayesian model selection improves causal discovery in complex datasets.
problem Identifying causal direction in Markov equivalence classes with realistic assumptions.
method Incorporating causal assumptions within Bayesian framework for model selection.
result Bayesian model selection outperforms previous methods on various datasets.
Learning new tasks with few samples using related task evaluations.
problem Learning a new task with limited data and related task evaluations.
method Modeling task relatedness through weak monotonicity and leveraging it in transfer learning and model selection aggregation.
result Pruning the model class based on monotonicity and hedging on the task frontier.
Despite the breakthroughs in accuracy and speed of single image super-resolution using faster and deeper convolutional neural networks, one central problem remains largely unsolved: how do we recover the finer texture details when we super-resolve at large upscaling factors? The behavior of optimization-based super-res…
We release a realistic, diverse, and challenging dataset for object detection on images. The data was recorded at a beer tent in Germany and consists of 15 different categories of food and drink items. We created more than 2,500 object annotations by hand for 1,110 images captured by a video camera above the checkout. …
BeGIN benchmarks GNNs for instance-dependent label noise in graphs.
problem Instance-dependent label noise in graph data.
method BeGIN introduces a benchmark with various noise types and evaluates noise-handling strategies across GNN architectures.
result Challenges of instance-dependent noise, especially LLM-based corruption, and the importance of node-specific parameterization.
PredictionMarketBench benchmarks trading agents on prediction markets.
problem Evaluating trading agents on prediction markets with realistic conditions.
method Deterministic replay of historical data, execution-realistic simulator, agent interface.
result Fee-aware algorithmic strategies outperform naive agents in volatile episodes.
The paper quantifies uncertainty in aggregated machine learning metrics.
problem Uncertainty in summarizing model performance across multiple tasks.
method Statistical methodologies including bootstrapping and Bayesian modeling.
result Insights into model performance dominance for specific tasks.
Paper recommends baselines for comparing GP approximations.
problem Lack of clear recommendations for comparing GP approximations.
method Developed a training procedure for the variational method of Titsias [2009].
result Variational method is a strong baseline that meets the recommended specification.
Unsupervised domain adaptation is a promising way to generalize deep models to novel domains. However, the current literature assumes that the label distribution is domain-invariant and only aligns the feature distributions or vice versa. In this work, we explore the more realistic task of Class-imbalanced Domain Adapt…
TSAM predicts directed temporal links using GCN and self-attention.
problem Predicting links in directed temporal networks.
method GCN, self-attention mechanism, autoencoder architecture, graph attentional layers, graph convolutional layers, graph recurrent unit layer.
result TSAM outperforms benchmarks on four realistic networks.
Semi-supervised learning (SSL) provides a powerful framework for leveraging unlabeled data when labels are limited or expensive to obtain. SSL algorithms based on deep neural networks have recently proven successful on standard benchmark tasks. However, we argue that these benchmarks fail to address many issues that th…
Generates synthetic manufacturing data for causal discovery benchmarking.
problem Lack of suitable real data for validating causal discovery algorithms.
method Distributional random forests for estimating conditional distributions.
result Semisynthetic manufacturing data adheres to a causal model.
CausalGame benchmarks LLM agents' causal thinking in games.
problem Evaluating causal thinking in AI Scientists with LLMs.
method Interactive games with 14 scenarios incorporating selection bias, measurement error, and hidden confounders.
result None of the 30 LLM agents demonstrated reliable causal thinking, with the best model achieving only 68.0% survival.