Theoretical model for iterative user discovery in recommender systems.
problem Iterative feedback loops in recommender systems and their biases.
method Theoretical framework to model system evolution and convergence properties.
result Theoretical bounds and convergence properties on user discovery and blind spots.
Last-iterate guarantees for learning in co-coercive games under noisy feedback.
problem Learning in co-coercive games with noisy feedback.
method Vanilla stochastic gradient descent with a new noise model.
result Last-iterate bound of order O(log(t)/t1/3) for co-coercive games. Efficient algorithm converges to Nash equilibrium in bilinear problems with bandit feedback.
problem Learning dynamics in bilinear saddle-point problems with bandit feedback.
method Uncoupled learning algorithm combining experimental design and FTRL with a tailored regularizer.
result Last-iterate convergence rate of ildeO(T−1/4) in high probability. 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.
We provide an online RLHF workflow for large language models.
problem Lack of online RLHF workflow for large language models.
method Construct preference models using open-source datasets, approximate human feedback, and implement online iterative RLHF.
result Our LLM achieves state-of-the-art performance on various benchmarks.
New algorithms converge faster to Nash equilibrium in zero-sum games with bandit feedback.
problem Learning in zero-sum games with bandit feedback without communication.
method Developed two uncoupled algorithms achieving optimal rate of Ω(T−1/4). result Achieved optimal rate of Ω(T−1/4) for convergence of policy profiles to Nash equilibrium. 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.
Mutation improves FTRL convergence in zero-sum games.
problem Lack of last-iterate convergence in FTRL variants.
method Introduced mutation to perturb action probabilities in FTRL.
result M-FTRL converges to Nash equilibria under full-information feedback.
A new one-point feedback scheme improves ZO algorithms for black-box optimization.
problem Optimizing black-box functions without gradient information.
method Proposes a one-point feedback scheme to estimate gradients using residuals.
result Matches query complexity of two-point schemes for deterministic Lipschitz functions.
This paper deals with bandit online learning problems involving feedback of unknown delay that can emerge in multi-armed bandit (MAB) and bandit convex optimization (BCO) settings. MAB and BCO require only values of the objective function involved that become available through feedback, and are used to estimate the gra…
New method prevents RLHF alignment collapse by accounting for policy's influence on reward model updates.
problem Iterative RLHF leads to alignment collapse where policies exploit RM's blind spots.
method Foresighted policy optimization (FPO) restores missing steering term via regularization.
result FPO prevents alignment collapse on LLM alignment pipelines using Llama-3.2-1B.
Improved text-to-image alignment using iterative VQA feedback.
problem Misalignment between text prompts and generated images, especially for complex inputs.
method Decompose complex prompts into assertions, evaluate each using VQA, combine scores iteratively.
result Significantly higher correlation with human ratings compared to CLIP, BLIP scores.
Stochastic structured prediction under bandit feedback follows a learning protocol where on each of a sequence of iterations, the learner receives an input, predicts an output structure, and receives partial feedback in form of a task loss evaluation of the predicted structure. We present applications of this learning …
PILAF optimizes reward models from human feedback for better policy alignment.
problem Creating accurate reward models from human feedback for policy optimization.
method Policy-Interpolated Learning for Aligned Feedback (PILAF) that explicitly aligns preference learning with maximizing underlying oracle reward.
result PILAF is optimal from both optimization and statistical perspectives, demonstrating strong performance in RLHF settings.
A new method for RLHF using proximal point Nash learning.
problem Capturing real human preferences in RLHF.
method Proximal point Nash learning, embedding self-play updates into a proximal point framework.
result High-probability last-iterate convergence for the combined method.
IDS improves RLHF by smoothing reward data, enhancing model performance.
problem Reward model performance degrades and overoptimization hinders true objective.
method Iterative Data Smoothing (IDS) updates model and data labels during each epoch.
result IDS outperforms traditional methods in RLHF.
AutoStan improves Bayesian models via predictive feedback.
problem Improving Bayesian models written in Stan.
method Iterative improvement of Stan models using NLPD and sampler diagnostics feedback.
result AutoStan can autonomously improve diverse Bayesian models across various structures.
The current leading computer vision models are typically feed forward neural models, in which the output of one computational block is passed to the next one sequentially. This is in sharp contrast to the organization of the primate visual cortex, in which feedback and lateral connections are abundant. In this work, we…
New algorithm for active learning from feedback coding.
problem Efficiently selecting examples for labeling in active machine learning.
method Formalized structural similarities between active learning and feedback channel coding, developed an optimal transport-based algorithm called Approximate Posterior Matching (APM).
result Learning performance comparable to existing methods at reduced computational cost.
Two coding schemes designed for different settings are compared.
problem Comparing two coding schemes designed for different communication settings.
method Demonstrated through evaluation of Deepcode and Modulo-SK on the AWGN channel with coded feedback and unit delay.
result Deepcode outperforms Modulo-SK in settings with coded feedback and noisy feedback.
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.
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.
Proposes RPG-RT for red-teaming T2I models without internal access.
problem Evaluating T2I models' security through red-teaming is challenging due to their closed-source nature and unknown defense mechanisms.
method Integrates LLM and rule-based preference modeling to dynamically adapt to unknown defense mechanisms.
result Demonstrates superior and practical approach for red-teaming T2I models.
Safe RL with binary feedback using SABRE algorithm.
problem Safe reinforcement learning with binary safety feedback.
method SABRE algorithm, combining active learning and reinforcement learning.
result Provable safe policy with high probability, no unsafe actions during training.
MAXMINLCB optimizes unknown target functions with preference feedback using a Stackelberg game approach.
problem Optimizing unknown target functions with pairwise comparisons and human feedback.
method MAXMINLCB, a zero-sum Stackelberg game, balances exploration and exploitation.
result MAXMINLCB consistently outperforms existing algorithms with a rate-optimal regret guarantee.
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.
FP uses random projections to train networks without feedback, achieving comparable performance to backpropagation.
problem Training neural networks without feedback from downstream layers.
method Forward Projection (FP) method that uses randomised nonlinear projections and closed-form regression.
result FP achieves comparable generalisation to backpropagation methods with a single forward pass, offering significant speedup.
SLHF uses sequential game theory to optimize preferences from human feedback.
problem Optimizing preferences from human feedback in sequential settings.
method SLHF frames the problem as a sequential-move game between Leader and Follower, decomposing the optimization into refinement and adversarial optimization.
result SLHF achieves strong alignment across diverse preference datasets and scales to large models.
Paper addresses RLHF alignment challenges with novel algorithms.
problem Challenges in RLHF alignment, especially in strategic exploration.
method Develops a reverse-KL regularized contextual bandit formulation and proposes efficient algorithms with theoretical guarantees.
result Proposed methods significantly outperform existing RLHF algorithms in real-world experiments.
One way to solve lasso problems when the dictionary does not fit into available memory is to first screen the dictionary to remove unneeded features. Prior research has shown that sequential screening methods offer the greatest promise in this endeavor. Most existing work on sequential screening targets the context of …
Interactive weak supervision learns useful heuristics from user feedback.
problem Creating useful heuristics for large labeled datasets is tedious and subjective.
method Develops an interactive framework for learning heuristics from user feedback.
result Only a few feedback iterations are needed to train models without ground truth labels.
The paper addresses learner privacy in convex optimization with feedback.
problem Privacy risks from eavesdropping adversaries observing learner's queries.
method Optimally obfuscating learner's queries to make their learned optimal value hard to estimate.
result Query complexity overhead is additive in L in the minimax formulation, multiplicative in L in the Bayesian formulation. ZSPO optimizes RL from unknown link functions using human feedback.
problem Designing RLHF algorithms for unknown link functions.
method Zero-order policy optimization with human preference feedback.
result ZSPO converges to a stationary policy with a polynomial rate.
Combines BO with context to optimize binary feedback.
problem Optimizing expensive binary functions with context.
method Bayesian active learning and optimization.
result Efficiently chooses best context and parameters.
Infrastructure monitors AI/ML radiology models across multiple sites.
problem Monitoring and improving AI/ML radiology models across multiple sites.
method Interactive radiology reporting, centralized cloud system, post-marketing surveillance.
result Efficient monitoring and iterative development of AI/ML models without radiologist burden.
The paper tackles learning from imperfect human feedback, especially in dueling bandit problems.
problem Learning from human feedback that can be irrational or imperfect.
method Developed a Robustified Stochastic Mirror Descent for Imperfect Dueling (RoSMID) algorithm.
result Achieved nearly optimal regret for dueling bandit problems under imperfect human feedback.
First sample-efficient algorithm for learning EFCE in bandit feedback settings.
problem Learning EFCE in bandit feedback settings for IIEFGs.
method Proposed K-EFCE and uncoupled no-regret algorithm with wide-range regret minimization. result First sample-efficient algorithm for learning EFCE from bandit feedback.
New method for RLHF reduces costs by integrating new data in one pass.
problem Continuous integration and re-optimization of models in RLHF leads to high computational and storage costs.
method Proposes a one-pass reward modeling method using online mirror descent with a tailored local norm.
result Achieves constant-time updates per iteration, enhancing both statistical and computational efficiency.
Data extracted from software repositories is used intensively in Software Engineering research, for example, to predict defects in source code. In our research in this area, with data from open source projects as well as an industrial partner, we noticed several shortcomings of conventional data mining approaches for c…
We consider stochastic multi-armed bandit problems with graph feedback, where the decision maker is allowed to observe the neighboring actions of the chosen action. We allow the graph structure to vary with time and consider both deterministic and Erdős-Rényi random graph models. For such a graph feedback model, we fir…
ESM-CNN uses error feedback to build a random CNN for time series forecasting.
problem Improving time series forecasting accuracy with CNNs.
method Incrementally adding random filters and neurons to adaptively construct a CNN.
result ESM-CNN outperforms state-of-the-art models in prediction accuracy and efficiency.
A new approach to fine-tuning LLMs with human feedback.
problem Inability of current reward models to fully represent human preferences.
method Introducing NLHF, a new pipeline for LLM fine-tuning using pairwise human feedback.
result NLHF produces a sequence of policies converging to the regularized Nash equilibrium.
New RLHF framework handles general preference oracles without reward functions.
problem Handling general preference oracles without assuming a reward function.
method Developed a minimax game between two LLMs for RLHF under a general preference oracle, focusing on KL-regularized preference.
result Proposed algorithms for efficient offline and online RLHF learning.
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.
BCCP uses bandit feedback to provide reliable predictions with limited labeled data.
problem Limited labeled data and bandit feedback challenge online set-valued classification.
method BCCP uses stochastic gradient descent to train model and make set-valued inferences with unbiased estimation of true label.
result BCCP offers coverage guarantees on a class-specific granularity.
AGFN improves causal discovery by integrating expert feedback and handling latent confounding.
problem Inaccurate causal discovery due to unreliable expert knowledge and latent confounding.
method Ancestral GFlowNet (AGFN) is a reinforcement learning algorithm that iteratively refines a policy based on noisy expert feedback to infer ancestral graphs.
result AGFN converges to the true ancestral graph given accurate expert responses and outperforms baselines in structural Hamming distance and Bayesian Information Criterion.
LLMs optimize quantum circuits by iteratively improving proposals with feedback and memory traces.
problem Optimizing quantum circuits using large language models (LLMs) under black-box evaluation.
method Closed-loop, test-time optimization with LLMs, score-difference feedback, and restart-from-the-best sampling.
result The approach improves circuit synthesis performance and success rate, especially for larger qubit settings.
Policy gradient method proves convergence in imperfect-information games.
problem Policy gradient methods in imperfect-information games (EFGs).
method Policy gradient approach with best-iterate convergence.
result Policy gradient leads to provable best-iterate convergence in self-play EFGs.