Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

16324763 · Jun 202019922001200920172026
48 results for top-two answers

Paper recovers top-two answers and confusion probability in multi-choice crowdsourcing.

problem Recovering top-two answers and confusion probability in multi-choice crowdsourcing tasks.
method Proposes a two-stage inference algorithm based on a model quantifying task difficulty and worker reliability.
result Achieves minimax optimal convergence rate and outperforms other algorithms in synthetic and real data experiments.

New method shows multi-objective bandits are not harder than single-objective ones.

problem Comparing multi-objective bandits to single-objective ones.
method Upper and lower confidence-bound estimators for every arm-objective pair, using top-two races and uncertainty-greedy rule.
result Achieves Pareto regret of \(O( icefrac{\log T}{g^\dagger})\), matching lower bound of \(Ω( icefrac{\log T}{g^\dagger})\).

We investigate and provide new insights on the sampling rule called Top-Two Thompson Sampling (TTTS). In particular, we justify its use for fixed-confidence best-arm identification. We further propose a variant of TTTS called Top-Two Transportation Cost (T3C), which disposes of the computational burden of TTTS. As our …

2019-10-24abs ↗pdf ↗

Unified algorithm for efficient pure exploration using dual variables.

problem Efficiently achieving a specific goal through adaptive experimentation.
method Introducing dual variables to derive optimal allocation conditions, leading to Information-Directed Selection.
result Top-two Thompson sampling attains asymptotic optimality for Gaussian best-arm identification.

BCI system improves word selection efficiency using sequential best-arm identification.

problem Conventional non-adaptive BCI paradigms lead to a lengthy learning process.
method Casted as sequence of best-arm identification tasks in multi-armed bandits, using pre-trained LLMs and STTS algorithm.
result Substantial empirical improvement in word selection efficiency demonstrated.

The paper studies privacy-protected BAI with fixed confidence, deriving lower bounds and proposing an adaptive algorithm.

problem Privacy-protected Best Arm Identification (BAI) in data-sensitive applications.
method Derives lower bounds on sample complexity, proposes AdaP-TT algorithm with Laplace noise, and validates with experiments.
result AdaP-TT matches the sample complexity lower bound up to constants in the high-privacy regime.

We consider a notion of balanced metrics for triples (X,L,E) which depend on a parameter α, where X is smooth complex manifold with an ample line bundle L and E is a holomorphic vector bundle over X. For generic choice of α, we prove that the limit of a convergent sequence of balanced metrics leads to a Hermitian-Einst…

2011-11-11abs ↗pdf ↗

The study uses unsupervised machine learning to identify top European football teams.

problem Selecting teams for the new European football Super League.
method Used Laplacian eigenmaps clustering on performance data.
result Successfully identified four clusters of teams based on performance metrics.

Unified model optimizes experiment performance and reduces duration.

problem Balancing reward maximization and experiment termination.
method Unified model that considers both within-experiment and post-experiment outcomes.
result Familiar algorithms can optimize a broad class of objectives with proper parameter adjustment.

EB-TCε identifies the best arm with ε confidence in stochastic bandits.

problem Identifying the best arm in stochastic bandits with a fixed level of confidence.
method EB-TCε is a novel sampling rule for ε-best arm identification in stochastic bandits.
result EB-TCε is the first anytime algorithm for fixed confidence or fixed budget identification.

Study adaptive sensing of Cox processes using posterior sampling and positive bases.

problem Adaptive sensing of Cox point processes with intensity function modeling.
method Model intensity function as truncated Gaussian process in positive basis, use Langevin dynamics and posterior sampling.
result Demonstrated improved sensing compared to classical Bayesian experimental design.

Power iteration has been generalized to solve many interesting problems in machine learning and statistics. Despite its striking success, theoretical understanding of when and how such an algorithm enjoys good convergence property is limited. In this work, we introduce a new class of optimization problems called scale …

2019-05-23abs ↗pdf ↗

Unified framework for ranking-and-selection with multiple correct answers and non-answerable estimates

problem Fixed-precision ranking-and-selection in structured settings with non-unique answers and non-answerable estimates
method Unified framework based on answer-wise acceptance sets, restricted generalized likelihood ratio stopping, and answer-pitfall decomposition
result Unified recipe performs well across a broad range of pure-exploration problems

A new algorithm identifies one of several nearly optimal arms in linear bandits.

problem Identifying one arm that is close to the best arm in linear bandits.
method Developed a procedure to adapt best-arm identification algorithms for ε\varepsilon-best-answer identification in transductive linear stochastic bandits.
result Proposed an asymptotically optimal algorithm for ε\varepsilon-best-answer identification.

The paper optimizes LLM accuracy by stopping early based on consistent answers.

problem Improving LLM accuracy in math and reasoning problems.
method Bayesian stopping policy to save on sampling costs, tracking only the L-1 most frequent answer counts.
result The L=3 stopping policy is sufficient for asymptotic optimality and significantly reduces inference costs.

We determine the sample complexity of pure exploration bandit problems with multiple good answers. We derive a lower bound using a new game equilibrium argument. We show how continuity and convexity properties of single-answer problems ensures that the Track-and-Stop algorithm has asymptotically optimal sample complexi…

2019-02-09abs ↗pdf ↗

Spectral methods improve signal recovery in mixed GLMs with precise asymptotics.

problem Estimating multiple signals from unlabeled observations in mixed GLMs.
method Developed exact asymptotics for spectral methods in a proportional regime.
result Optimized spectral method combined with a linear estimator minimizes estimation error.

Sampling more can make models more confident in wrong answers, not better.

problem The modal ceiling and correlation ceiling limit the benefit of increased sampling.
method Analyzes the trade-offs between sampling more and selecting the best answer.
result Extra sampling beyond a certain point does not improve model performance and can even degrade it.

Adaptive algorithm identifies best arm with abstention, showing phase transition from polynomial to exponential error probability.

problem Bayesian best-arm identification with abstention to reduce undetected error.
method Adaptive algorithm PGWS that optimally uses abstention budget.
result Introducing any positive abstention budget induces an exponential decay in undetected error probability.

Paper analyzes biases in video QA datasets, showing models can answer 37-48% questions correctly without multimodal context.

problem Question answering biases in video QA datasets can lead to model overfitting and poor generalization.
method Analyzed popular video question answering datasets, conducted ablation studies on biases from annotators and question types.
result Pretrained language models can answer 37-48% questions correctly without multimodal context, far exceeding random guess baseline.

UCB algorithm's arm-sampling behavior is revealed, leading to new insights and proofs.

problem Optimizing multi-armed bandit algorithms for worst-case scenarios.
method Analysis of UCB algorithm's arm-sampling behavior and process-level characterization.
result UCB's arm-sampling rates are asymptotically deterministic, regardless of problem complexity.

FinanceBench benchmarks LLMs on financial QA, revealing limitations.

problem Evaluating LLMs' performance on financial question answering.
method Developed a comprehensive test suite (FinanceBench) with 10,231 questions, tested 16 models, and manually reviewed answers.
result Existing LLMs have significant limitations for financial QA, especially GPT-4-Turbo.

Extends uncertainty detection in neural networks to finer distinctions.

problem Detecting finer distinctions between certain, uncertain, and out-of-distribution points.
method Two-step approach: first builds class distribution using Kernel Activation Vectors, second determines test point confidence.
result Corrects overconfident NN decisions and learns to say 'I don't know' when uncertain.

A novel method for visual question answering using scene graphs and reinforcement learning.

problem Answering free-form questions about images with deep linguistic and visual understanding.
method Context-driven, sequential reasoning based on scene graphs and reinforcement learning.
result Our method almost reaches human performance on the GQA dataset.