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

3673109145 · May 202619922001200920172026
48 results for feedback shift

A method corrects feedback shift in predicting conversion rates with delayed feedback.

problem Delayed feedback leads to mislabeling of positive instances in training data.
method Uses importance weight approach to correct feedback shift.
result Proposed method outperforms existing methods in offline and online experiments.

Algorithm minimizes regret in predictive models influenced by their own predictions.

problem Finding near-optimal models under performativity with unknown shifts.
method Developed an algorithm that uses performative feedback to achieve low regret, scaling only with distribution shift complexity.
result Achieved regret bounds scaling with distribution shift complexity, not reward function complexity.

The paper tackles exact linearization and control of flat discrete-time systems.

problem Exact linearization and control of flat nonlinear discrete-time systems.
method Investigates conditions for choosing new inputs and feedbacks that may depend on forward-shifts of the new input.
result Easily verifiable conditions for choosing a feasible input and a new input that minimizes forward-shifts of the flat output.

New approach mitigates feedback divergence in imitation learning.

problem Divergence between held-out error and learner performance in imitation learning.
method Identifies covariate shift as the root cause and proposes a simulator-based solution.
result Naive behavioral cloning performs well in real-world decision making problems.

Theory for RLHF generalization under reward shift and clipped KL.

problem Theoretical understanding of RLHF generalization, especially with reward shift and clipped KL.
method Developed generalization theory for RLHF, accounting for reward shift and clipped KL.
result Presented generalization bounds for RLHF, suggesting generalization error from sampling, reward shift, and KL clipping.

Study uses geometric algebra to analyze credit cycles, revealing dangerous feedback loops.

problem Understanding and predicting dangerous feedback loops in credit cycles.
method Represent economic states as multi-vectors in Clifford algebra, focusing on bivector elements for rotational coupling.
result Geometric relationship between unemployment and credit contraction shifts from simple correlation to dangerous rotational dynamics during crises.

We study the adversarial multi-armed bandit problem where partial observations are available and where, in addition to the loss incurred for each action, a \emph{switching cost} is incurred for shifting to a new action. All previously known results incur a factor proportional to the independence number of the feedback …

2019-07-29abs ↗pdf ↗

Not all types of supervision signals are created equal: Different types of feedback have different costs and effects on learning. We show how self-regulation strategies that decide when to ask for which kind of feedback from a teacher (or from oneself) can be cast as a learning-to-learn problem leading to improved cost…

2019-07-11abs ↗pdf ↗

CausalRM models rewards from user feedback, overcoming noise and bias.

problem Aligning language models with user preferences from noisy, biased feedback.
method Causal-theoretic reward modeling framework addressing noise and bias in observational feedback.
result CausalRM learns accurate reward signals from noisy and biased observational feedback.

Paper tackles RLHF with DCPPO method, proving near-optimal suboptimality.

problem Challenges in offline RLHF with limited human feedback and bounded rationality.
method DCPPO method involving three stages: MLE, reward function recovery, and pessimistic value iteration.
result DCPPO's suboptimality almost matches classical pessimistic offline RL in terms of distribution shift and dimension.

The paper tackles performative policy learning with strategic agents, improving scalability and generalizability.

problem Strategic agents adjust their features in response to a released policy, causing endogenous distribution shifts.
method Relaxing parametric assumptions, the paper uncovers a low-dimensional structure in distribution shifts and proposes a gradient-based policy optimization algorithm.
result The proposed algorithm achieves high sample efficiency and provides theoretical guarantees for convergence.

Partially performative prediction studies how predictive models influence future data.

problem Distribution shift in predictive models due to endogenous and exogenous factors.
method Generalizing performative prediction to capture both endogenous and exogenous sources of distribution shift.
result Developed online analogues of performative stability and optimality for partially performative environments.

Study on tracking preference shifts in dueling bandits problems.

problem Tracking significant preference shifts in dueling bandits problems.
method Analysis of dueling bandits with distribution shifts, focusing on significant shifts (Suk and Kpotufe, 2022).
result Design of adaptive algorithms with O(KildeLT)O(\sqrt{K ilde{L}T}) dynamic regret for certain preference distribution classes.

ECI improves time series prediction uncertainty quantification by smoothing miscoverage error.

problem Challenges in uncertainty quantification for time series prediction due to temporal dependence and distribution shift.
method Error-quantified Conformal Inference (ECI) by smoothing quantile loss function and introducing adaptive feedback scale.
result ECI achieves valid miscoverage control and tighter prediction sets than existing methods.

New framework ensures valid uncertainty estimates for any data stream changes.

problem Challenges of distribution shifts and adversarial actors in real-world data streams.
method Leveraging Blackwell approachability from game theory, the framework guarantees calibrated uncertainties for any compact space.
result Improves calibration and decision-making for energy systems.

ADPO optimizes relative advantage in reinforcement learning from human feedback.

problem Optimizing policy alignment in reinforcement learning from human preferences.
method ADPO explicitly parameterizes the optimal structure through anchored logits, decoupling response quality from prior popularity.
result Empirically, ADPO achieves state-of-the-art performance on reasoning tasks, outperforming GRPO by 30.9 percent.

Paper proposes methods to help autonomous vehicles adapt to unexpected driving scenarios.

problem Autonomous vehicles struggle with unexpected driving conditions.
method Robust imitative planning (RIP) and adaptive robust imitative planning (AdaRIP) methods to detect and adapt to distribution shifts.
result Methods outperform current state-of-the-art approaches in nuScenes prediction challenge.

SAIL improves online alignment of large language models with minimal feedback.

problem Offline RLHF methods often lead to sub-optimal performance due to fixed preference datasets.
method SAIL uses bilevel optimization and a single-level first-order method to iteratively refine model alignment.
result SAIL significantly improves alignment performance on open-sourced datasets with minimal computational overhead.

Paper develops methods to optimize policies directly from human feedback without reward inference.

problem Challenges in RLHF, including reward model overfitting and distribution shift.
method Develops two algorithms for RLHF without reward inference, using zeroth-order gradient approximators.
result Establishes polynomial convergence rates and outperforms existing methods in numerical experiments.

New method for efficient online exploration in RLHF reduces regret.

problem Efficiently collecting new preference data in RLHF to refine reward model and policy.
method Proposes a new exploration scheme that directs preference queries toward reducing uncertainty in reward differences most relevant to policy improvement.
result Establishes regret bounds of order T(β+1)/(β+2)T^{(β+1)/(β+2)} for online RLHF, with polynomial scaling in all model parameters.

New RLHF algorithm identifies optimal policies from human feedback without explicit reward inference.

problem Training large language models with human feedback without reward inference.
method Model-free RLHF algorithm BSAD\mathsf{BSAD} that identifies optimal policies directly from human preference.
result Provable, instance-dependent sample complexity ildeO(cMSA3H3Mlog1δ) ilde{\mathcal{O}}(c_{\mathcal{M}}SA^3H^3M\log\frac{1}δ).

New framework for predicting decisions that influence their own outcomes.

problem Predictions that affect the outcomes they predict, leading to undesirable distribution shift.
method Risk minimization framework combining statistics, game theory, and causality.
result Necessary and sufficient conditions for retraining to converge to a performatively stable point of minimal loss.

Online L2D algorithm for multiclass classification with varying experts.

problem Handling streaming data, changing expert availability, and shifting expert distribution.
method First online L2D algorithm with O((n+ne)T2/3)O((n+n_e)T^{2/3}) and O((n+ne)T)O((n+n_e)\sqrt{T}) regret guarantees.
result Effective extension of standard L2D to settings with varying expert availability and reliability.

A trading system uses LLMs to adapt to volatile crypto markets.

problem Volatility and market sentiment in cryptocurrencies make traditional models ineffective.
method Specialized LLM agents for technical analysis, sentiment evaluation, and decision-making; verbal feedback for continuous improvement.
result Agents outperform buy-and-hold strategy with consistent gains across market phases.

Robots' agility in changing terrain helps financial models adapt to market shifts.

problem Challenges in financial market forecasting due to regime switching.
method Adapts pretrained LLMs using intrinsic market rewards and reinforcement learning.
result Significantly improved accuracy in adapting to market regime shifts.

Active learning improves inspection systems by using weakly labeled data.

problem Rapidly updating machine vision inspection systems in evolving manufacturing processes.
method Developed a methodology for active learning from weakly labeled data, addressing covariate shift with domain-adversarial training.
result Demonstrated that active learning can accelerate the annotation process and reduce false positives.

Efficiently poisons offline RLHF models by flipping preference labels.

problem Vulnerability of offline RLHF models to preference label flipping attacks.
method Developed two attack methods: BAL-A and BMP-A, solving a structured binary sparse approximation problem.
result Demonstrated that flipping one preference label induces a parameter-independent shift in the DPO gradient, enabling structured binary sparse approximation.

Adaptive ML learns complex time-varying systems without new data.

problem Applying ML to time-varying systems with shifting distributions.
method Mapping high-dimensional inputs to low-dimensional latent space, actively tuning latent space based on feedback.
result Learning correlations and tracking system evolution in real-time without new data.

New framework formalizes RLHF trilemma: improving safety, fairness, and robustness is computationally infeasible.

problem Aligning large language models with diverse human values while maintaining computational feasibility and robustness.
method Complexity-theoretic analysis integrating statistical learning theory and robust optimization.
result Achieving both representativeness (epsilon <= 0.01) and robustness (delta <= 0.001) for global-scale populations requires super-polynomial operations.

New insights into cascade feedback linearization of control systems.

problem Obtaining a cascade feedback linearization for invariant control systems.
method Introducing truncated versions of operators from the calculus of variations to prove new theorems.
result Established new geometry and foundational theorems for future work.

New method learns from either positive or negative feedback alone.

problem Limited applicability of existing preference optimization methods in scenarios with only unpaired feedback.
method Decouples learning from positive and negative feedback, using expectation-maximization (EM) to optimize probability of positive outcomes and explicitly incorporate negative examples.
result Stable learning from negative feedback alone demonstrated.

Study model selection in batch policy optimization with three error sources.

problem Learn a policy competitive with the best model class in batch policy optimization.
method Formalized in contextual bandit setting with linear model classes, addressing approximation error, statistical complexity, and dataset shift.
result No algorithm can optimally trade-off all three error sources, but relaxing any one enables near-oracle inequalities for the others.

User preferences for items can be inferred from either explicit feedback, such as item ratings, or implicit feedback, such as rental histories. Research in collaborative filtering has concentrated on explicit feedback, resulting in the development of accurate and scalable models. However, since explicit feedback is oft…

2011-09-27abs ↗pdf ↗

This paper improves image retrieval accuracy through novel relevance feedback methods.

problem Improving image retrieval accuracy in Content-Based Image Retrieval (CBIR).
method Novel addition to feature re-weighting and classification techniques, focusing on 0-th iteration improvement.
result Significantly improved retrieval accuracy from relevance feedback.