Revisits PPO design choices, exposing failure modes and proposing alternatives.
problem Failure modes of standard PPO in new environments.
method Revisits standard PPO design choices, exposes failure modes, and proposes alternative approaches.
result Alternative design choices prevent failure modes in new environments.
GBS uses machine learning to design products based on consumer preferences.
problem Designing products to meet consumer preferences.
method GBS is a discrete choice experiment that uses machine learning to adaptively construct paired comparison questions.
result GBS outperforms existing methods in accuracy and sample efficiency.
The study examines how experimental design choices affect machine learning model performance.
problem Lack of guidelines on choosing experimental designs and machine learning models.
method 12 experimental designs, 7 families of predictive models, 7 test functions, 8 noise settings.
result Guidelines for practical applications of DOE and ML are provided.
How can we design safe reinforcement learning agents that avoid unnecessary disruptions to their environment? We show that current approaches to penalizing side effects can introduce bad incentives, e.g. to prevent any irreversible changes in the environment, including the actions of other agents. To isolate the source…
Study investigates key design choices in on-policy RL algorithms.
problem Lack of transparency in RL algorithm implementations.
method Implemented >50 design choices in a unified RL framework.
result Insights and recommendations for on-policy RL training.
Alt-GNNs improve travel mode choice modeling by integrating graph neural networks with GEV models.
problem Capturing alternative dependence in discrete choice models with predefined, symmetric, and uniform dependence.
method Introducing Alternative Graph Neural Networks (Alt-GNNs) that embed alternative dependence within a unified framework.
result Alt-GNNs significantly improve predictive performance over benchmark models in travel mode choice datasets.
Human irrationality can improve AI design, study shows.
problem Improving AI by learning from human decision-making biases.
method Developed a novel POMDP model to simulate human decision-making in contextual choice tasks.
result Reinforcement learners can exploit human irrationalities to make better decisions.
Develops deep learning models for choice modeling.
problem Computational intractability and sample inefficiency in existing choice model learning methods.
method Deep learning-based choice models in two settings: feature-free and feature-based.
result Demonstrates improved recovery of existing choice models and reduced sample complexity.
Automates supervised learning pipeline design with matrix and tensor factorization.
problem Designing effective supervised learning pipelines with many choices.
method Uses matrix and tensor factorization to model pipeline search space and develops greedy experiment design protocols.
result Demonstrates the effectiveness of the approach on real-world classification problems.
The way that people make choices or exhibit preferences can be strongly affected by the set of available alternatives, often called the choice set. Furthermore, there are usually heterogeneous preferences, either at an individual level within small groups or within sub-populations of large groups. Given the availabilit…
AI framework for automated policy-making connects with econometrics and social choice.
problem Improving government and economic policy-making through AI.
method Social Environment Design framework integrating Reinforcement Learning, EconCS, and Computational Social Choice.
result Promotes ethical and responsible decision making by solving open problems.
Researchers dissect Neural ODEs to understand their dynamics.
problem Understanding the inner workings of Neural ODEs.
method Developing continuous-depth formulation to clarify design choices.
result Clarified the influence of design choices on Neural ODE dynamics.
Design of printed circuit board (PCB) stack-up requires the consideration of characteristic impedance, insertion loss and crosstalk. As there are many parameters in a PCB stack-up design, the optimization of these parameters needs to be efficient and accurate. A less optimal stack-up would lead to expensive PCB materia…
New DR algorithm preserves both local and global structure.
problem Trade-off between preserving local and global structure in DR methods.
method Analysis of existing DR methods and design principles for loss functions.
result Design of PaCMAP algorithm that preserves both local and global structure.
Whereas deep neural network (DNN) is increasingly applied to choice analysis, it is challenging to reconcile domain-specific behavioral knowledge with generic-purpose DNN, to improve DNN's interpretability and predictive power, and to identify effective regularization methods for specific tasks. This study designs a pa…
A recommendation framework helps users choose healthcare interventions.
problem Choice overload in online healthcare communities.
method Multi-Armed Bandit (MAB) approach with innovative model components.
result Our recommendation design outperforms state-of-the-art systems.
Unified perspective on natural gradient methods for GMMs, improving variational inference.
problem Efficiently learning multi-modal approximations of complex distributions.
method Comparison and optimization of VIPS and iBayes-GMM methods for Gaussian mixture models.
result Hybrid approach significantly outperforms both VIPS and iBayes-GMM.
Bayesian methods detect significant IIA violations in similarity choice data.
problem Detecting IIA violations in similarity choice data complicates classical models.
method Proposed two statistical methods: classical goodness-of-fit test and Bayesian PPC.
result Significant IIA violations confirmed in both datasets, driven by context effects.
We study mechanisms to characterize how the asymptotic convergence of backpropagation in deep architectures, in general, is related to the network structure, and how it may be influenced by other design choices including activation type, denoising and dropout rate. We seek to analyze whether network architecture and in…
This paper deals with the explicit design of strategy formulations to make the best strategic choices from a conventional matrix form of representing strategic choices. The explicit strategy formulation is an analytical model which is targeted to provide a mathematical strategy framework to find the best moment for str…
Automated HPO design using Bayesian optimization and benchmarking.
problem Designing effective hyperparameter optimization algorithms is manual and lacks systematic understanding.
method Formalized space of HPO candidates, Bayesian optimization for search, ablation analysis.
result Simple configurations can perform well in HPO, especially with right parameters.
Optimal adaptive experiment for choosing best treatment with binary outcomes.
problem Choosing the best treatment from binary options in an adaptive experiment.
method Adaptive experiment with two phases: treatment allocation and choice. Neyman allocation method used.
result Neyman allocation is minimax and Bayes optimal, matching lower bounds for regret.
Designs algorithms to assist humans without affecting their decisions.
problem Algorithms often fail to improve human decisions.
method Formalizes algorithm design using potential outcomes framework and monotonicity assumption.
result Derives minimax optimal recommendation algorithms for limited data.
Graph neural nets improve discrete choice modeling with network effects.
problem Modeling network effects in discrete choice problems.
method Graph Convolutional Neural Network (GCNN) architecture.
result Higher predictive performance than standard models with interpretability.
We introduce Bayesian optimization, a technique developed for optimizing time-consuming engineering simulations and for fitting machine learning models on large datasets. Bayesian optimization guides the choice of experiments during materials design and discovery to find good material designs in as few experiments as p…
Gaussian processes are ubiquitous in machine learning, statistics, and applied mathematics. They provide a flexible modelling framework for approximating functions, whilst simultaneously quantifying uncertainty. However, this is only true when the model is well-specified, which is often not the case in practice. In thi…
Unified diffusion framework enhances generative models flexibility.
problem Improving generative models' design freedom and efficiency.
method Unified framework incorporating choice of representation, prior distribution, and noise scheduling.
result Enhanced flexibility leading to more efficient training and data generation.
Semantic segmentation remains a computationally intensive algorithm for embedded deployment even with the rapid growth of computation power. Thus efficient network design is a critical aspect especially for applications like automated driving which requires real-time performance. Recently, there has been a lot of resea…
Bayesian DL model improves DCMs for better predictive and inferential performance.
problem Limited interpretability and predictive underperformance of traditional DCMs.
method Integrates deep learning with approximate Bayesian inference (SGLD).
result Improves predictive and inferential metrics in discrete choice models.
Researchers propose a new SSL risk decomposition method to evaluate and improve self-supervised learning models.
problem Self-supervised learning evaluation is limited to a single metric, providing little insight into model performance and improvement.
method Proposes an SSL risk decomposition that considers four error components: approximation, representation usability, probe generalization, and encoder generalization.
result Analysis of 169 SSL vision models reveals the main sources of error and provides insights for improving SSL models in specific settings.
New formulations capture aversion to ambiguity about volatility.
problem Capturing aversion to ambiguity about unknown and time-varying volatility.
method Introduces novel preference formulations and compares them with existing models.
result Illustrates the impact of ambiguity aversion in static and dynamic models.
Utilizing recently introduced concepts from statistics and quantitative risk management, we present a general variant of Batch Normalization (BN) that offers accelerated convergence of Neural Network training compared to conventional BN. In general, we show that mean and standard deviation are not always the most appro…
We investigate two new strategies for the numerical solution of optimal stopping problems within the Regression Monte Carlo (RMC) framework of Longstaff and Schwartz. First, we propose the use of stochastic kriging (Gaussian process) meta-models for fitting the continuation value. Kriging offers a flexible, nonparametr…
We use learning curves to analyze deep networks and evaluate model design.
problem Evaluate design choices in deep networks.
method Propose a method to robustly estimate learning curves, abstract their parameters, and evaluate different parameterizations.
result Interesting observations on the effectiveness of different parameterizations.
New probability path model improves flow matching forecasting performance.
problem Impact of probability path model selection on flow matching forecasting performance.
method Proposed a novel probability path model designed to improve forecasting performance.
result Our model achieves faster convergence during training and improved predictive performance compared to existing models.
Gaussians as noise in NCE lead to exponentially bad conditioning, hindering its efficiency.
problem Exponential conditioning of Hessian in NCE with Gaussian noise.
method Using Gaussian as the noise distribution in NCE.
result Gaussian noise in NCE leads to exponentially bad conditioning of the loss Hessian.
Studies across many disciplines have shown that lexical choice can affect audience perception. For example, how users describe themselves in a social media profile can affect their perceived socio-economic status. However, we lack general methods for estimating the causal effect of lexical choice on the perception of a…
We consider the problem of identifying the most profitable product design from a finite set of candidates under unknown consumer preference. A standard approach to this problem follows a two-step strategy: First, estimate the preference of the consumer population, represented as a point in part-worth space, using an ad…
AgensFlow learns multi-agent coordination policies from experience.
problem Difficult coordination choices in multi-agent systems built on LLMs.
method Online policy learning from repeated trajectories, treating decisions as learnable.
result Learned routing improves coordination-heavy workflows over static wiring.
The task of image generation started to receive some attention from artists and designers to inspire them in new creations. However, exploiting the results of deep generative models such as Generative Adversarial Networks can be long and tedious given the lack of existing tools. In this work, we propose a simple strate…
This paper presents the current state of a work in progress, whose objective is to better understand the effects of factors that significantly influence the performance of Latent Semantic Analysis (LSA). A difficult task, which consists in answering (French) biology Multiple Choice Questions, is used to test the semant…
RLHF performs well despite violating social choice theory axioms.
problem RLHF's empirical success contradicts social choice theory axioms.
method Showed RLHF satisfies pairwise majority and Condorcet consistency under mild assumptions, and introduced new alignment criteria.
result RLHF satisfies pairwise majority and Condorcet consistency under mild assumptions, explaining its practical success.
Grale designs graphs for graph learning, improving performance on large datasets.
problem Finding the right graph for semi-supervised learning with billions of nodes.
method Fuses multiple similarity measures using locality sensitive hashing to create task-specific graphs.
result Grale detects a large number of malicious actors, increasing recall by 89%.
Unified framework for machine learning interatomic potentials.
problem Designing and optimizing machine learning models for interatomic potentials.
method Unified mathematical framework unifying ACE and NequIP, providing a systematic design space.
result Demonstrated through ablation studies, critical design choices for high accuracy.
The paper explores various forms of calibration scores and their implications for fairness.
problem The evaluation of probabilistic predictions through calibration.
method The authors organize three grouping choices and one agglomeration of group errors, providing a framework for comparing and creating new calibration scores.
result The study demonstrates that appropriate choices of grouping can provide notions of (sub-)group or individual fairness.
Bayesian optimisation's mean function choice affects convergence speed.
problem The choice of mean function in Bayesian optimisation impacts convergence speed.
method Empirical investigation of 8 mean functions on 10 synthetic and 2 real-world problems.
result Using a constant mean function equal to the worst observed quality value promotes faster convergence.
New model improves website ranking by considering user choices as a whole.
problem Optimizing content ordering for user clicks in website design.
method Introduced multinomial logit (MNL) choice model to LTR framework, proposing UCB algorithms.
result Proved theoretical bounds on regret for UCB algorithms in both known and unknown position parameter settings.
We address feature interpretation and reproducibility issues in dense nets, proposing a modified loss function.
problem Feature interpretation and reproducibility issues in dense nets.
method Proposed a modified loss function to circumvent basis collapse.
result Substantially concise nets with 100x fewer parameters and lower MSE loss.