New algorithm for minimizing regret in adversarial dueling bandits.
problem Minimizing regret in dueling bandits with adversarial feedback.
method Introduced an algorithm with -round regret of .
result Algorithm achieves regret lower bound.
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.
168,657 papers · 148 categories
Trend · papers per month
48 results for “Adversarial Preferences”
New algorithm optimizes dueling bandits for both stochastic and adversarial preferences.
problem Optimizing decision-making in environments where only relative preferences are observed.
method Proposed a reduction from dueling bandits to multi-armed bandits, achieving optimal regret bounds.
result First best-of-both-world result for dueling bandits, optimal regret bound for Condorcet-winner benchmark.
New framework for reinforcement learning with adversarial preferences in tabular MDPs.
problem Learning from preferences rather than direct rewards in MDPs with adversarial settings.
method Developed PbMDPs framework, established lower bounds, and proposed algorithms for regret minimization.
result Achieved regret bounds of for PbMDPs with Borda scores.
Paper tackles regret bounds and exploration complexity for multi-objective reinforcement learning with picky preferences.
problem Formalizing multi-objective reinforcement learning with adversarial preferences.
method Model-based algorithm with nearly optimal regret bound and preference-free exploration.
result Achieves nearly minimax optimal regret bound and nearly optimal trajectory complexity.
A new, computationally friendly formula for a class of risk-averse preferences.
problem Characterizing a class of risk-averse preferences called uniformly weighted divergence preferences.
method Introducing a new formula that characterizes UWDP as the translation-invariant hull of state-independent expected utility.
result UWDP are the translation-invariant hull of state-independent expected utility over .
Provably Mitigating Overoptimization in RLHF: Your SFT Loss is Implicitly an Adversarial Regularizercs.LG
Mitigates overoptimization in RLHF by reformulating SFT loss as a preference optimization loss.
problem Overoptimization in RLHF where reward model misguides generative model.
method Proposes a theoretical algorithm that minimizes maximum likelihood estimation and reward penalty, reformulates as simple objective combining preference optimization and supervised learning losses.
result Improved performance of RPO compared to DPO baselines in aligning LLMs.
We introduce two tactics to attack agents trained by deep reinforcement learning algorithms using adversarial examples, namely the strategically-timed attack and the enchanting attack. In the strategically-timed attack, the adversary aims at minimizing the agent's reward by only attacking the agent at a small subset of…
Paper shows RLHF can be solved similarly to standard RL.
problem Difficulty of RLHF compared to standard RL.
method Reduction to reward-based RL techniques.
result RLHF can be solved using existing algorithms for reward-based RL.
Matching Markets meet Cumulative Prospect Theory: Towards Optimal and Adversarially Robust Learningcs.LG
Matching Markets meet Cumulative Prospect Theory: Towards Optimal and Adversarially Robust Learning
problem Multi-agent multi-armed bandit problem in competitive setup with two-sided matching markets under human-centric decision making model
method Using cumulative prospect theory (CPT) to emulate human preferences
result Improved regret guarantees in adversarial markets with CPT as risk-sensitive measure
CnGAN generates synthetic user preferences for non-overlapped users in cross-network recommender systems.
problem Cross-network recommender solutions ignore non-overlapped users, limiting their applicability.
method Multi-task learning, encoder-GAN architecture, user-based pairwise loss function.
result Generated user preferences improve recommendations for non-overlapped users, achieving superior performance.
Recommender systems aim to find an accurate and efficient mapping from historic data of user-preferred items to a new item that is to be liked by a user. Towards this goal, energy-based sequence generative adversarial nets (EB-SeqGANs) are adopted for recommendation by learning a generative model for the time series of…
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.
SRPO improves AI alignment with human preferences through self-improvement and task-independent optimization.
problem AI models trained with RLHF lack self-correction mechanisms and struggle with task generalization.
method SRPO formulates the preference learning problem as a min-max objective, optimizing a self-improvement policy and a generative policy in an adversarial fashion, making the solution task-independent.
result SRPO outperforms existing methods, achieving 90% AI Win-Rate on XSum and 56% on Arena-Hard prompts after a single revision.
The goal of task transfer in reinforcement learning is migrating the action policy of an agent to the target task from the source task. Given their successes on robotic action planning, current methods mostly rely on two requirements: exactly-relevant expert demonstrations or the explicitly-coded cost function on targe…
A new model for sequential prediction handles adversarial examples by allowing abstention.
problem Sequential prediction algorithms fail with adversarial examples, leading to incorrect predictions.
method Proposes a new model that allows abstention from predictions on adversarial examples, scaling error with VC dimension.
result A learner's error scales with the VC dimension of the hypothesis class, matching the stochastic setting.
AugmentedPCA improves PCA with supervised or adversarial objectives.
problem Lack of reproducible linear analogs for deep latent factor models.
method Augments PCA with supervised or adversarial objectives.
result Improves downstream classification performance and identifies cancer-related genes.
Adaptive algorithms minimize regret in matching markets with contextual arm preferences.
problem Minimizing regret in matching markets with context-dependent player utilities.
method Developed adaptive algorithms for stochastic and adversarial contexts, providing upper and lower bounds.
result Achieved sublinear regret bounds for both stochastic and adversarial contexts.
In recent years, the Word2Vec model trained with the Negative Sampling loss function has shown state-of-the-art results in a number of machine learning tasks, including language modeling tasks, such as word analogy and word similarity, and in recommendation tasks, through Prod2Vec, an extension that applies to modeling…
The task of image generation started to receive some attention from artists and designers to inspire them in new creations. However, exploiting the results of deep generative models such as Generative Adversarial Networks can be long and tedious given the lack of existing tools. In this work, we propose a simple strate…
In this paper we establish rigorous benchmarks for image classifier robustness. Our first benchmark, ImageNet-C, standardizes and expands the corruption robustness topic, while showing which classifiers are preferable in safety-critical applications. Then we propose a new dataset called ImageNet-P which enables researc…
Despite the great achievements of the modern deep neural networks (DNNs), the vulnerability/robustness of state-of-the-art DNNs raises security concerns in many application domains requiring high reliability. Various adversarial attacks are proposed to sabotage the learning performance of DNN models. Among those, the b…
Paper generalizes strategic classification framework and introduces SVC for PAC-learning.
problem Strategic manipulation of testing data to fool classifiers.
method Unified framework for strategic classification, strategic VC-dimension (SVC).
result Characterizes the learnability and computational tractability of linear classifiers.
Paper analyzes robustness of non-Lipschitz networks, proving powerful adversarial attacks but offering solutions.
problem Adversarial attacks on deep networks, especially non-Lipschitz networks.
method Developed an attack model that abstracts the challenge of adversarial robustness, proving the power of such attacks and offering solutions.
result Proves powerful adversarial attacks on non-Lipschitz networks but offers solutions with abstention.
We consider learning of fundamental properties of communities in large noisy networks, in the prototypical situation where the nodes or users are split into two classes according to a binary property, e.g., according to their opinions or preferences on a topic. For learning these properties, we propose a nonparametric,…
Paper tackles ranking items with a semi-random comparison graph and a monotone adversary.
problem Ranking items based on pairwise comparisons from a semi-random comparison graph with a monotone adversary.
method Developed a weighted maximum likelihood estimator (MLE) and an SDP-based approach to reweight the semi-random graph.
result Achieves near-optimal sample complexity, up to a log^2(n) factor, for identifying the top-K preferred items.
A new method uses counterfactual learning to improve recommendation system evaluation.
problem Inconsistent results in recommender systems due to exposure mechanisms.
method Proposes a minimax empirical risk formulation with an adversarial game to account for exposure.
result Shows improved learning bounds and effectiveness over various recommendation settings.
Composite Travel Generative Adversarial Networks for Tabular and Sequential Population Synthesiscs.LG
CTGAN synthesizes population data for travel behavior simulation.
problem Synthesizing population data for agent-based transportation modeling.
method Composite Travel Generative Adversarial Network (CTGAN).
result Consistent and accurate generation of synthetic populations with tabular and sequential mobility data.
Reducing the Probability of Undesirable Outputs in Language Models Using Probabilistic Inferencecs.LG
RePULSe improves language model alignment by reducing undesired outputs without sacrificing overall performance.
problem Aligning language models with human preferences while minimizing undesired outputs.
method Integrates probabilistic inference into RL training to reduce undesired outputs.
result RePULSe achieves a better balance between expected reward and undesired output probability.
One pixel modification can make deep models unlearnable.
problem Protecting data from unauthorized training of deep neural networks.
method Perturbing only one pixel in each image to degrade model accuracy.
result Generated One-Pixel Shortcut (OPS) cannot be erased by adversarial training and strong augmentations.
Optimizes molecular generation for chemist preferences.
problem Models lack inherent preferences for chemist-desired structures.
method Fine-tuning with Direct Preference Optimization.
result Approach is simple, efficient, and highly effective.
Paper argues the bear case for Bitcoin is bounded and terminal states are neutral to positive.
problem The identity of Bitcoin's creator and the associated overhang risk.
method Quantitative analysis of Satoshi's 1.148 million BTC position, considering various preference sets.
result The terminal states most consistent with observed behavior are neutral to slightly positive for Bitcoin's effective supply.
CoExBO optimizes lithium-ion batteries with user input, enhancing trust and efficiency.
problem User distrust in Bayesian optimization due to opacity and lack of user input.
method Preference learning and iterative explanation to integrate user insights.
result Algorithm converges to optimal solution even with adversarial user inputs.
New method adapts to user preferences dynamically, improving recommendation models.
problem Current recommendation models lack dynamic adaptation to changing user preferences.
method Preference Discerning with LLM-Enhanced Generative Retrieval
result Mender achieves state-of-the-art performance in adapting to evolving user preferences.
Many real-world engineering problems rely on human preferences to guide their design and optimization. We present PrefOpt, an open source package to simplify sequential optimization tasks that incorporate human preference feedback. Our approach extends an existing latent variable model for binary preferences to allow f…
Enhances preference learning by incorporating response times into binary choices.
problem Limited information from binary choices about preference strength.
method Combines choices and response times using the EZ diffusion model.
result Response times improve utility estimation for strong preferences.
Bayesian optimization learns DM preferences for multi-outcome experiments.
problem Optimizing expensive experiments with unknown utility functions and multiple outcomes.
method Alternates preference learning and Bayesian optimization, using pairwise comparisons.
result Preference exploration strategies improve Bayesian optimization performance.
New study shows personalized content recommendations can lead to polarization of user preferences.
problem Personalized content recommendations can alter user preferences, leading to polarization.
method Used a model of preference dynamics to explore how personalized content affects user preferences.
result Standard reward maximization algorithms achieve only constant regret in personalized recommendation environments.
This work proves MultiKrum is robust in mean estimation with adversaries.
problem Mean estimation in the presence of Byzantine adversaries.
method Introducing κ* and constructing upper and lower bounds on MultiKrum's robustness coefficient.
result MultiKrum is the first provably robust aggregation rule, with robustness coefficient bounds.
Bayesian approach quantifies uncertainty in LLM evaluations.
problem Statistical uncertainty in evaluating LLM behavior.
method Bayesian evaluation of LLM behavior using probabilistic text generation strategies.
result Bayesian approach provides useful uncertainty quantification about LLM behavior.
Bayesian optimization agent learns user preferences from pairwise comparisons.
problem Learning user preferences from unknown and infinite choices.
method Sequential Bayesian optimization with pairwise comparisons.
result Optimal agent strategy minimizes remaining system uncertainty.
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.
This paper studies robust forward investment and consumption preferences within a zero-volatility context. Different from previous works, we consider an incomplete financial market model due to general investment portfolio constraints. We provide a new PDE characterization and a novel semi-explicit saddle-point constru…
In preference-based reinforcement learning (RL), an agent interacts with the environment while receiving preferences instead of absolute feedback. While there is increasing research activity in preference-based RL, the design of formal frameworks that admit tractable theoretical analysis remains an open challenge. Buil…
Study on identifying most preferred policy in bandits with vector-valued rewards.
problem Identifying the most preferred policy in bandits with vector-valued rewards.
method Derive a novel lower bound on sample complexity, design the Preference-based Track and Stop (PreTS) algorithm, and derive a new concentration inequality.
result The sample complexity of PreTS is asymptotically tight.
We study the top- ranking problem where the goal is to recover the set of top- ranked items out of a large collection of items based on partially revealed preferences. We consider an adversarial crowdsourced setting where there are two population sets, and pairwise comparison samples drawn from one of the populat…
Stable and consistent model alignment for language models without assuming human preference models.
problem Lack of statistical consistency in existing alignment methods.
method Relative density ratio optimization between preferred and mixture of preferred and non-preferred data distributions.
result Our approach achieves statistical consistency and stability, providing tighter convergence guarantees.
Dropping a tiny fraction of preferences can significantly alter the rankings of top LLMs.
problem Robustness of LLM ranking systems to small changes in preference data.
method A computational method based on the Bradley-Terry model to evaluate robustness.
result Top LLM rankings can be highly sensitive to the removal of a small fraction of preferences.
Paper explores limits and possibilities of aligning LLMs with human preferences.
problem Aligning LLMs with diverse human preferences to ensure fairness and informed outcomes.
method Analysis of probabilistic representation of human preferences and preservation of diverse preferences.
result LLMs can't fully align with human preferences using reward-based approaches due to Condorcet cycles, but mixed strategies are statistically possible.