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

285684112 · Jun 202019922001200920172026
48 results for Feedback Divergence

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.

This work develops a unified framework for RLHF with general ff-divergence regularization.

problem Theoretical understanding of general ff-divergence regularization in RLHF.
method Holistic approach across ff-divergence class, two algorithms based on distinct sampling principles.
result Provably efficient algorithms with O(logT)O(\log T) regret and O(1/T)O(1/T) sub-optimality gap.

This paper introduces ff-DPO, a generalized approach to Direct Preference Optimization using diverse divergence constraints.

problem Aligning large language models with human preferences while mitigating safety risks.
method Incorporates diverse divergence constraints to simplify the relationship between reward and optimal policy, eliminating the need for estimating the normalizing constant.
result Optimizes LLMs to align with human preferences more efficiently and under a broader set of divergence constraints.

We consider the problem of learning from sparse and underspecified rewards, where an agent receives a complex input, such as a natural language instruction, and needs to generate a complex response, such as an action sequence, while only receiving binary success-failure feedback. Such success-failure rewards are often …

2019-02-19abs ↗pdf ↗

Paper proposes f-DPG for aligning language models with preferences.

problem Aligning language models with user preferences.
method Uses f-divergence to approximate target distributions and minimizes a forward KL from it using DPG.
result Jensen-Shannon divergence often outperforms forward KL divergence, leading to significant improvements.

We study the KK-armed dueling bandit problem, a variation of the standard stochastic bandit problem where the feedback is limited to relative comparisons of a pair of arms. We introduce a tight asymptotic regret lower bound that is based on the information divergence. An algorithm that is inspired by the Deterministic…

2015-06-08abs ↗pdf ↗

An optimal feedback controller for a given Markov decision process (MDP) can in principle be synthesized by value or policy iteration. However, if the system dynamics and the reward function are unknown, a learning agent must discover an optimal controller via direct interaction with the environment. Such interactive d…

2019-07-06abs ↗pdf ↗

Extends reinforcement learning alignment to scalar rewards, improving math reasoning.

problem Designing reinforcement learning algorithms for general LLM alignment.
method Introduces f-GRPO and f-HAL, estimating f-divergences between reward-aligned and unaligned distributions.
result Improves math reasoning RLVR tasks and mitigates reward hacking.

Model captures decision-making under bounded rationality with prior beliefs and market feedback.

problem Bounded rationality in decision-making with limited processing abilities.
method Maximum entropy principle applied to Quantal Response Statistical Equilibrium framework.
result Prior beliefs influence decision-making, altering the outcome of market feedback.

Neural networks are commonly trained to make predictions through learning algorithms. Contrastive Hebbian learning, which is a powerful rule inspired by gradient backpropagation, is based on Hebb's rule and the contrastive divergence algorithm. It operates in two phases, the forward (or free) phase, where the data are …

2018-06-19abs ↗pdf ↗

A capsule is a collection of neurons which represents different variants of a pattern in the network. The routing scheme ensures only certain capsules which resemble lower counterparts in the higher layer should be activated. However, the computational complexity becomes a bottleneck for scaling up to larger networks, …

2018-08-11abs ↗pdf ↗

Paper offers a fast convergence theory for offline decision making.

problem Offline decision making problems, including reinforcement learning and off-policy evaluation.
method Introduces a framework (DMOF) and algorithm (EDD) with a fast convergence guarantee.
result Demonstrates a fast convergence guarantee with a lower bound complement.

The success of popular algorithms for deep reinforcement learning, such as policy-gradients and Q-learning, relies heavily on the availability of an informative reward signal at each timestep of the sequential decision-making process. When rewards are only sparsely available during an episode, or a rewarding feedback i…

2018-05-25abs ↗pdf ↗

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.

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.

Develops a new model for RLHF accounting for partially observed states and intermediate feedback.

problem Lack of models for partially observed states and intermediate feedback in RLHF.
method PORRL model with cardinal and dueling feedback methods.
result Demonstrates improved learning and alignment with new model-based and model-free methods.

Classifier learns to ignore unreliable feedback from end users.

problem Improving classifier performance by filtering unreliable feedback.
method Modeling end users as autonomous agents, periodically retraining classifier with filtered feedback.
result Classifier can identify and filter out unreliable feedback, improving performance.

Advocates a local feedback approach for RL in unknown systems.

problem Finding optimal feedback laws in unknown nonlinear dynamical systems.
method Searches over a local feedback representation consisting of an open-loop sequence and an optimal linear feedback law.
result Results in highly efficient training and superior performance compared to global methods.

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.

Study how communication and feedback graphs affect learning outcomes.

problem Understanding the impact of feedback graphs on cooperative online learning.
method Analyzed network regret in terms of the independence number of the strong product of communication and feedback graphs.
result Proved bounds for network regret and demonstrated the non-improvable nature of positive results in pathological cases.

New method uses correlated auxiliary feedback to reduce regret in parameterized bandits.

problem Reducing regret in parameterized bandits with correlated auxiliary feedback.
method Develops a reward estimator using auxiliary feedback with tight confidence bounds.
result Shows significant reduction in regret compared to standard methods.

PolySwarm uses a swarm of LLMs to predict and arbitrage prediction markets.

problem Real-time prediction market trading and latency arbitrage inefficiencies.
method PolySwarm employs a swarm of 50 diverse LLMs, Bayesian combination, and risk-controlled execution.
result Swarm aggregation outperforms single-model baselines in prediction tasks.

The problem of feedback equivalence for control systems is considered. An algebra of differential invariants and criteria for the feedback equivalence for regular control systems are found.

2008-12-07abs ↗pdf ↗

Study of reinforcement learning with additional feedback observations.

problem Episodic reinforcement learning in Markov decision processes with feedback observations.
method Formalization of feedback graph, model-based algorithms leveraging feedback, regret bound analysis.
result Regret bound depends only on the size of the maximum acyclic subgraph of the feedback graph.

DRO-REBEL improves LLM alignment by robustly updating models online.

problem Overfitting and drifting of LLMs during RLHF.
method DRO-REBEL uses type-pp Wasserstein, KL, and χ2χ^2 ambiguity sets for robust online updates.
result DRO-REBEL achieves faster convergence and better performance than prior methods.

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 noisy bandit feedback for multiclass classification.

problem Learning multiclass classifier with corrupted feedback.
method Proposes an unbiased estimator technique to estimate noise rates and an end-to-end framework.
result Algorithm achieves mistake bounds of O(T)O(\sqrt{T}) in high noise and O(Ticefrac23)O(T^{ icefrac{2}{3}}) in worst case.

We propose a generalization of the best arm identification problem in stochastic multi-armed bandits (MAB) to the setting where every pull of an arm is associated with delayed feedback. The delay in feedback increases the effective sample complexity of standard algorithms, but can be offset if we have access to partial…

2018-03-29abs ↗pdf ↗

Active learning framework for optimizing human preferences in reinforcement learning.

problem Selecting most informative feedback for training models of human preferences.
method Proposes an active learning framework to collect preferential feedback online or offline.
result Errors in DPO logit estimates diminish with more feedback.

The paper optimizes portfolios using MACD signals derived from price history.

problem Optimizing risky asset portfolios with latent mean-reverting and momentum factors.
method Derives optimal strategies based on MACD signals from EMA processes.
result Establishes admissibility and verification of optimal strategies.

Paper studies CLO with partial feedback, improving decision-making in uncertain contexts.

problem Improving decision-making in contexts with uncertain cost coefficients using partial feedback.
method Unified class of offline learning algorithms for CLO with different types of feedback, using IERM framework.
result Fast-rate regret bound for IERM with partial feedback and misspecified model classes.

We use online convex optimization (OCO) for setpoint tracking with uncertain, flexible loads. We consider full feedback from the loads, bandit feedback, and two intermediate types of feedback: partial bandit where a subset of the loads are individually observed and the rest are observed in aggregate, and Bernoulli feed…

2017-09-12abs ↗pdf ↗