Optimizes query routing to LLMs under cost and resource constraints.
problem Non-uniform or adversarial batching in per-query routing methods leads to cost inefficiency.
method Batch-level, resource-aware routing framework that jointly optimizes model assignment for each batch.
result Robust routing framework improves accuracy by 1-14% over non-robust methods.
Efficiently learns reward functions with fewer queries and shorter computation times.
problem Expensive data generation and labeling in robot learning.
method Batch active preference-based learning methods using determinantal point processes (DPP) and heuristic alternatives.
result Our batch active learning algorithm requires only a few queries and computes them in a short amount of time.
Unified method for deep active learning improves performance and efficiency.
problem Improving deep active learning performance and efficiency.
method Unified and principled approach using Wasserstein distance for querying and training.
result Consistently better empirical performance and time-efficient query strategy compared to baselines.
Data generation and labeling are usually an expensive part of learning for robotics. While active learning methods are commonly used to tackle the former problem, preference-based learning is a concept that attempts to solve the latter by querying users with preference questions. In this paper, we will develop a new al…
BMBO-DARN optimizes expensive functions with varying fidelities.
problem Optimizing expensive, multi-fidelity functions efficiently.
method Batch Multi-fidelity Bayesian Optimization with Deep Auto-Regressive Networks.
result BMBO-DARN improves surrogate learning and optimization performance.
We propose a new batch mode active learning algorithm designed for neural networks and large query batch sizes. The method, Discriminative Active Learning (DAL), poses active learning as a binary classification task, attempting to choose examples to label in such a way as to make the labeled set and the unlabeled pool …
Meta-algorithm for efficient reinforcement learning from human preferences.
problem Learning from human preference comparisons in Markov decision processes.
method Randomized exploration and experimental design for batch comparison queries.
result Meta-algorithm achieves both regret and last-iterate guarantees with minimal preference queries.
In many high-throughput experimental design settings, such as those common in biochemical engineering, batched queries are more cost effective than one-by-one sequential queries. Furthermore, it is often not possible to directly choose items to query. Instead, the experimenter specifies a set of constraints that genera…
CRUMB: Efficient Prior Fitted Network Inference via Distributionally Matched Context Batching
problem Inference of tabular foundation models with large training datasets
method CRUMB (Clustered Retrieval Using Minimised-MMD Batching)
result CRUMB outperforms state-of-the-art context selection strategies on the TabArena benchmark
In many real-world machine learning applications, unlabeled data are abundant whereas class labels are expensive and scarce. An active learner aims to obtain a model of high accuracy with as few labeled instances as possible by effectively selecting useful examples for labeling. We propose a new selection criterion tha…
The study analyzes batched methods for early stopping in stochastic multi-armed bandits.
problem Early stopping in stochastic multi-armed bandits with fixed confidence.
method Instance-dependent lower bounds and a general batched algorithm with upper bounds.
result Upper and lower bounds on the number of batches and sample complexity.
Batch Thompson Sampling reduces exploration-exploitation trade-off in online decision making.
problem Balancing exploration and exploitation in online decision making.
method Introducing a batch Thompson Sampling framework for stochastic multi-arm bandit and linear contextual bandit problems.
result Achieves asymptotic regret bound with O(logT) batch queries, significantly reducing interactions. Paper addresses privacy and robustness in stochastic linear bandits.
problem Stochastic linear bandits with differential privacy and adversarial robustness.
method Logarithmic batch queries, arm elimination algorithm, two privacy models.
result First algorithms providing differential privacy and adversarial robustness.
New method reduces labeling costs in semi-supervised learning.
problem Efficiently label data with limited labeled samples.
method Formulates batch acquisition as bilevel optimization.
result Shows effectiveness in keyword detection tasks.
Active search is a learning paradigm for actively identifying as many members of a given class as possible. A critical target scenario is high-throughput screening for scientific discovery, such as drug or materials discovery. In this paper, we approach this problem in Bayesian decision framework. We first derive the B…
A new BO method tackles high-dimensional optimization without reconstruction.
problem Optimizing high-dimensional black-box functions is challenging, especially when low-dimensional structures are assumed.
method Tackles the problem in the original high-dimensional space using learned low-dimensional structure.
result Our method explores the high-dimensional space more effectively than existing approaches.
We present a generic framework for trading off fidelity and cost in computing stochastic gradients when the costs of acquiring stochastic gradients of different quality are not known a priori. We consider a mini-batch oracle that distributes a limited query budget over a number of stochastic gradients and aggregates th…
The paper tackles fast rates in batch active learning with pool-based data.
problem Reduced adaptivity in batch active learning leads to suboptimal results.
method Proposes a stage-wise greedy algorithm that balances informativeness and diversity.
result The algorithm's excess risk matches minimax rates in standard statistical learning settings.
Bayesian optimization (BO) has become an effective approach for black-box function optimization problems when function evaluations are expensive and the optimum can be achieved within a relatively small number of queries. However, many cases, such as the ones with high-dimensional inputs, may require a much larger numb…
New algorithm guarantees performance on noisy data.
problem Learning with noisy data and heavy-tailed distributions.
method Anytime online-to-batch conversion for smooth objectives.
result Stochastic gradient-based algorithm with sub-Gaussian error bounds.
Differentiable learning via SGD and GD can simulate various learning problems, depending on precision and minibatch size.
problem Understanding the power of differentiable learning via SGD and GD compared to statistical query (SQ) learning.
method Comparing the learning power of SGD and GD on population and empirical losses with statistical query learning.
result The learning power of SGD and GD depends on the precision of gradient calculations relative to the minibatch size or sample size.
BOE reformulates BO as a classifier for scalable batch optimisation.
problem Scalable batch optimisation of expensive functions.
method Reformulates BO as density-ratio estimation, removing need for explicit function prior.
result Theoretical guarantees and improved uncertainty estimates for batch optimisation.
Leveraging the wealth of unlabeled data produced in recent years provides great potential for improving supervised models. When the cost of acquiring labels is high, probabilistic active learning methods can be used to greedily select the most informative data points to be labeled. However, for many large-scale problem…
In many real-world AD applications including computer security and fraud prevention, the anomaly detector must be configurable by the human analyst to minimize the effort on false positives. One important way to configure the detector is by providing true labels (nominal or anomaly) for a few instances. Recent work on …
Data collection and labeling is one of the main challenges in employing machine learning algorithms in a variety of real-world applications with limited data. While active learning methods attempt to tackle this issue by labeling only the data samples that give high information, they generally suffer from large computa…
sGPO optimizes RLVR training by balancing inference FLOPs and training efficiency.
problem RLVR training allocates rollout budget without considering query difficulty.
method sGPO trades inference FLOPs for reduced training FLOPs.
result sGPO matches or exceeds baseline performance while reducing training compute by a factor of three.
In this paper, we consider the convex and non-convex composition problem with the structure n1∑i=1nFi(G(x)), where G(x)=n1∑j=1nGj(x) is the inner function, and Fi(⋅) is the outer function. We explore the variance reduction based met…
Adapts data analysis for growing data, improving generalization guarantees.
problem Challenges of overfitting and statistical validity in adaptive workflows with growing data.
method Generalizes adaptive analysis on dynamic data, incorporating time-varying empirical accuracy bounds and mechanisms.
result First generalization bounds for adaptive analysis on dynamic data, matching prior works' improvement over data splitting.
Study learns random hypergraphs with queries, improving on previous results.
problem Learn random hypergraphs with non-adaptive queries.
method Equivalence to group testing, using Erdős-Rényi model for graphs.
result Generalization to random k-uniform hypergraphs. Stochastic momentum methods trade compute efficiency for serial runtime.
problem Stochastic momentum methods trade compute efficiency for serial runtime.
method Stochastic HB and ASGD for consistent linear regression with Gaussian covariates.
result HB preserves SGD-level CE over a larger batch-size window, allowing larger batches to reduce serial runtime until HB reaches its deterministic accelerated scale.
Data compression is a popular technique for improving the efficiency of data processing workloads such as SQL queries and more recently, machine learning (ML) with classical batch gradient methods. But the efficacy of such ideas for mini-batch stochastic gradient descent (MGD), arguably the workhorse algorithm of moder…
We develop an algorithm for minimizing a function using n batched function value measurements at each of T rounds by using classifiers to identify a function's sublevel set. We show that sufficiently accurate classifiers can achieve linear convergence rates, and show that the convergence rate is tied to the difficu…
This work uses RL to optimize batch experiments in SDOE.
problem Maximizing knowledge with limited resources and constraints.
method Reinforcement Learning for batch-sampling in Bayesian SDOE.
result The proposed algorithm optimizes experiment selection for multiple tasks.
SOBER optimizes and quadrates efficiently in parallel for diverse tasks.
problem Scalability of batch Bayesian optimization and quadrature for expensive functions.
method Reformulates batch selection as a quadrature problem, balancing exploitation and exploration.
result SOBER outperforms 11 baselines on 12 tasks.
Unified study of active learning for deep neural networks.
problem Improving performance of deep neural networks through active learning.
method Investigation of incremental and cumulative training modes, model configurations, query strategies, and pseudo-labels.
result Proposed more efficient querying procedures and insights into active learning behavior.
TES optimizes black-box functions efficiently with minimal approximations.
problem Efficient Bayesian optimization with minimal approximations and generalization to batch BO.
method TES acquisition function measures information gain on trusted maximizers.
result TES achieves state-of-the-art performance with minimal approximations.
Gradient descent fails to learn simple neural networks efficiently.
problem Learning one-layer neural networks efficiently using gradient descent.
method Gradient descent and statistical query algorithms.
result Superpolynomial lower bounds for learning one-layer neural networks.
ε-shotgun optimizes expensive function evaluations in parallel.
problem Optimizing expensive black-box functions efficiently.
method ε-greedy batch Bayesian optimisation.
result ε-shotgun outperforms state-of-the-art batch methods.
Estimates TV distance between autoregressive models under different access models.
problem Estimating the total variation distance between two autoregressive distributions.
method Three access models: sample access, logit access, and noisy logit access; provides query complexity for each.
result Improved query complexity for estimating TV distance in autoregressive models.
AdvMind detects adversary intent in black-box attacks with high accuracy.
problem Detecting adversary intent in black-box adversarial attacks is challenging.
method AdvMind accounts for adversary adaptiveness and synthesizes queries to expose intent.
result AdvMind detects adversary intent with over 75% accuracy after observing less than 3 query batches.
QActor optimizes learning from noisy labeled data streams by querying experts for clean labels.
problem Learning from noisy labeled data in continuous streams with limited oracle queries.
method Combines quality models for filtering and oracle queries for true labels, dynamically adjusting query limits.
result QActor nearly matches optimal accuracy with up to 6% additional ground truth data from experts.
Optimization of high-dimensional black-box functions is an extremely challenging problem. While Bayesian optimization has emerged as a popular approach for optimizing black-box functions, its applicability has been limited to low-dimensional problems due to its computational and statistical challenges arising from high…
Heterogeneous network embedding (HNE) is a challenging task due to the diverse node types and/or diverse relationships between nodes. Existing HNE methods are typically unsupervised. To maximize the profit of utilizing the rare and valuable supervised information in HNEs, we develop a novel Active Heterogeneous Network…
A new framework reduces traffic congestion by 36%.
problem Efficient traffic signal control in a metropolitan area.
method Two-stage framework combining Evolution Strategies and multi-agent reinforcement learning.
result Reduces traffic congestion by 36%.
Unified analysis of first-order methods for smooth games using IQCs.
problem Certify convergence rates of first-order methods for smooth and strongly-monotone games.
method Adapted integral quadratic constraints (IQCs) to study first-order methods and derive tight upper bounds of convergence rates.
result First global convergence rate for the negative momentum method with O(κ1.5) iteration complexity. New analysis for black-box learning without gradients, improving generalization bounds.
problem Generalization error analysis for derivative-free optimization.
method Zeroth-order Stochastic Search (ZoSS) algorithm for Lipschitz and smooth losses.
result Generalization bounds independent of model dimension, batch size, and number of perturbed evaluations.
Influence functions help study large language model generalization, revealing surprising decay patterns.
problem Understanding and mitigating risks in large language models (LLMs).
method Eigenvalue-corrected Kronecker-Factored Approximation (EK-FAC) to scale influence functions to LLMs.
result Influences decay to near-zero when key phrases order is flipped, revealing a surprising limitation.
Service robots benefit from encoding information in semantically meaningful ways to enable more robust task execution. Prior work has shown multi-relational embeddings can encode semantic knowledge graphs to promote generalizability and scalability, but only within a batched learning paradigm. We present Incremental Se…