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.

168,695 papers · 148 categories

Trend · papers per month

54108162216 · Jun 202019922001200920172026
48 results for policy violations

Measures policy-violating content prevalence with ML-assisted sampling and LLM labeling.

problem Accurate measurement of content violations that are often rare and costly to label.
method Design-based measurement system using ML-assisted probability sampling and LLM labeling.
result Produces unbiased prevalence estimates with confidence intervals and dashboard drilldowns.

Canary optimizes VaR-constrained RL problems with a conservative bound using Cantelli's inequality.

problem Optimizing reinforcement learning policies under VaR constraints in dense cost regimes.
method Employing Cantelli's inequality to create a conservative and smooth bound on VaR constraints based on moments of cost returns. Extending trust-region framework for worst-case bounds on policy improvement and constraint violation.
result Canary reliably satisfies VaR constraints with fewest violations and earliest permanent satisfaction, while maintaining reward competitiveness.

New algorithm reduces sample complexity for safe reinforcement learning.

problem Safe reinforcement learning in constrained MDPs with performance and safety constraints.
method Model-based primal-dual algorithm balancing regret and bounded constraint violations.
result Proves near-optimal policies with bounded violations or zero violations in CMDPs.

Study optimizes portfolio allocation policies using off-policy data and constraints.

problem Optimizing portfolio allocation policies under constraints using off-policy data.
method Solves a minimax objective with off-policy estimators and online learning to control constraint violations.
result Constructs near-optimal allocation policies for various regimes of operation and constraints.

VaR-CPO optimizes VaR-constrained RL problems with conservative policy updates.

problem Optimizing VaR-constrained reinforcement learning problems.
method Combines Cantelli's inequality and trust-region framework for efficient and conservative optimization.
result Achieves zero constraint violations during training in feasible environments.

Paper improves COCO problem, reducing constraint violation at the cost of slightly more regret.

problem Online Convex Optimization with adversarial constraints.
method Proposes new policies that trade off regret for reduced constraint violation.
result Achieves ildeO(dT+Tβ) ilde{O}(\sqrt{dT}+ T^β) regret and ildeO(dT1β) ilde{O}(dT^{1-β}) CCV.

DOLCE improves off-policy evaluation and learning by decomposing effects.

problem Bias in off-policy evaluation and learning due to policy mismatch.
method Uses lagged contexts and a moment-based training procedure to decompose and cancel bias.
result DOLCE achieves substantial improvements in off-policy evaluation and learning.

This paper detects Markov violations in RL with noise, improving policy development.

problem Partial observability and sensor/actuator noise invalidate Markovian assumptions in RL.
method Combines PCMCI causal discovery with Markov Violation score (MVS).
result Even substantial noise doesn't always disrupt multi-step dependencies.

We propose a method to optimise the parameters of a policy which will be used to safely perform a given task in a data-efficient manner. We train a Gaussian process model to capture the system dynamics, based on the PILCO framework. Our model has useful analytic properties, which allow closed form computation of error …

2017-12-15abs ↗pdf ↗

FISAR uses neural networks to optimize safe reinforcement learning with forward-invariant constraints.

problem Safe reinforcement learning with constraints in safety-critical environments.
method Imposing linear constraints on policy parameters' updating dynamics, using a DNN-based optimizer to satisfy these constraints.
result The policy decreases constraint violation and maximizes cumulative reward monotonically.

Algorithm safely learns from sub-optimal baseline policies while satisfying constraints.

problem Safe reinforcement learning with constraints when baseline policy is sub-optimal.
method Iterative policy optimization alternating between return maximization, baseline distance minimization, and constraint projection.
result Consistently outperforms baselines, achieving 10x fewer constraint violations and 40% higher reward.

FOCOPS optimizes agent's behavior while adhering to constraints.

problem Optimizing agent's behavior while respecting safety constraints.
method FOCOPS solves a constrained optimization problem in policy space, then projects the solution back into the parametric space.
result FOCOPS achieves better performance on constrained robotics tasks.

We propose a novel approach to address one aspect of the non-stationarity problem in multi-agent reinforcement learning (RL), where the other agents may alter their policies due to environment changes during execution. This violates the Markov assumption that governs most single-agent RL methods and is one of the key c…

2019-11-28abs ↗pdf ↗

New method estimates policy performance under unobserved confounding.

problem Estimating policy performance when decisions depend on unobserved variables.
method Developed worst-case bounds for robust OPE under unobserved confounding.
result Efficient procedure for computing worst-case bounds, proving statistical consistency.

Paper develops a method to estimate value of a policy in confounded MDPs.

problem Estimating value of a policy in the presence of unmeasured confounders.
method Uses auxiliary variables to identify target policy's value in a confounded MDP.
result Develops an off-policy value estimator robust to model misspecification.

While learning in an unknown Markov Decision Process (MDP), an agent should trade off exploration to discover new information about the MDP, and exploitation of the current knowledge to maximize the reward. Although the agent will eventually learn a good or optimal policy, there is no guarantee on the quality of the in…

2020-02-08abs ↗pdf ↗

New RL algorithm achieves sublinear regret and constraint violation without simulators.

problem Maximizing reward under utility constraints in large-scale systems.
method Model-free, simulator-free algorithm using LSVI-UCB with primal-dual optimization and soft-max policy.
result Achieves ildeO(d3H3T) ilde{\mathcal{O}}(\sqrt{d^3H^3T}) regret and ildeO(d3H3T) ilde{\mathcal{O}}(\sqrt{d^3H^3T}) constraint violation bounds.

This paper studies the evaluation of policies that recommend an ordered set of items (e.g., a ranking) based on some context---a common scenario in web search, ads, and recommendation. We build on techniques from combinatorial bandits to introduce a new practical estimator that uses logged data to estimate a policy's p…

2016-05-16abs ↗pdf ↗

Risk-controlled post-processing optimizes decision policies under risk constraints.

problem Optimizing decision policies with risk constraints for better outcomes.
method Developed a post-processing algorithm that selects a threshold based on fitted fallback policy and score, leveraging tools from algorithmic stability and stochastic processes.
result The post-processed policy achieves precise expected risk control under exchangeability and meets or nearly meets risk budgets while preserving more agreement with the baseline.

VERAFI improves financial AI by verifying calculations and compliance.

problem Financial AI systems generate errors and violations during reasoning.
method VERAFI combines dense retrieval, reranking, and automated reasoning policies.
result VERAFI achieves 94.7% factual correctness, 81% relative improvement.

New framework improves restless bandit policies for large numbers of arms.

problem Efficiently compute policies for large numbers of arms in restless bandit problems.
method Follow-the-Virtual-Advice framework, converting single-armed policies to N-armed policies.
result Achieves an O(1/\sqrt{N}) optimality gap in both discrete and continuous settings.

In distributed reinforcement learning, it is common to exchange the experience memory of each agent and thereby collectively train their local models. The experience memory, however, contains all the preceding state observations and their corresponding policies of the host agent, which may violate the privacy of the ag…

2019-07-15abs ↗pdf ↗

The paper analyzes CMDPs, balancing exploration and exploitation to avoid constraint violations.

problem Balancing exploration and exploitation in CMDPs to satisfy constraints.
method Two approaches: optimistic planning and incremental updates of primal and dual variables.
result Both approaches achieve sublinear regret on utility and constraint violations, with stronger guarantees for the linear programming approach.

A new estimator reduces variance in slate bandit OPE.

problem Large action spaces in slate bandits cause high variance in OPE.
method Develops Latent IPS (LIPS) to optimize slate abstractions for low variance and bias.
result LIPS substantially outperforms existing estimators in scenarios with non-linear rewards and large slate spaces.

The options framework is a popular approach for building temporally extended actions in reinforcement learning. In particular, the option-critic architecture provides general purpose policy gradient theorems for learning actions from scratch that are extended in time. However, past work makes the key assumption that ea…

2019-12-31abs ↗pdf ↗

A new method improves actor-critic RL by integrating HMC, enhancing policy distribution and exploration.

problem Actor-critic RL yields suboptimal policies due to amortization gap and insufficient exploration.
method Integrating Hamiltonian Monte Carlo (HMC) into the actor-critic RL framework.
result Improves policy distribution and exploration, leading to better policy estimates and higher returns.

Unified market making controls risk, arbitrage, and volatility surfaces.

problem Market making risk, arbitrage, and volatility surface consistency.
method Constrained RL and stochastic control for risk-sensitive execution and hedging.
result Agent achieves positive P&L with zero calendar and butterfly violations.

Online learning with one-sided feedback aims to maximize accuracy while ensuring fairness.

problem Maximizing accuracy in online learning with limited feedback and ensuring fairness.
method Extending the framework of Bechavod et al. (2020) to incorporate dynamic panels of auditors, reducing the problem to a contextual combinatorial semi-bandit, and leveraging Exp2 and Context-Semi-Bandit-FTPL algorithms.
result Multi-criteria no regret guarantees for accuracy and fairness are provided.

This work highlights problems with off-policy estimation in recommender systems due to unobserved confounders.

problem Evaluation of recommender systems under unobserved confounders.
method Policy-based estimators and characterisation of statistical bias due to confounding.
result Naive propensity estimation under confounding leads to severely biased metric estimates.

The paper addresses fairness in dynamic pricing for strategic buyers.

problem Price disparities among specific groups can lead to unfair perceptions and legal violations.
method Proposes a dynamic pricing policy that achieves fairness and discourages strategic behavior.
result Achieves an upper bound of O(T+H(T))O(\sqrt{T}+H(T)) regret over TT time horizons, reducing regret by 35.06% compared to a benchmark policy.

The paper reveals that baselines significantly impact RL algorithms' convergence.

problem Understanding the true impact of baselines on policy optimization.
method Theoretical analysis of bandit and RL problems, focusing on natural policy gradient and EXP3.
result Baselines can determine algorithm convergence, contradicting traditional optimization theory.

StepMix algorithm ensures safe exploration in reinforcement learning with near-optimal performance.

problem Conservative exploration in reinforcement learning under episode-wise constraints.
method StepMix algorithm balances exploitation and exploration while ensuring episode-wise conservative constraint.
result StepMix achieves near-optimal regret order as in the constraint-free setting.

This paper provides theoretical foundations for using quantized actions in behavior cloning.

problem Applying autoregressive models to continuous control requires discretizing actions through quantization, which is poorly understood.
method The paper analyzes quantization error propagation and statistical sample complexity, and proposes model-based augmentation.
result Behavior cloning with quantized actions achieves optimal sample complexity, matching existing lower bounds.