BeBold improves exploration in sparse-reward tasks by regulating visitation counts.
problem Efficient exploration in deep reinforcement learning under sparse rewards.
method Regulated difference of inverse visitation counts.
result BeBold solves 12 challenging tasks in MiniGrid with fewer steps than previous state-of-the-art.
RECODE uses clustering and embedding to track state visitation counts in RL.
problem Efficient novelty-based exploration in nonstationary RL environments.
method Non-parametric clustering, online density estimation, inverse dynamics loss.
result RECODE achieves state-of-the-art performance in challenging RL tasks.
Study geodesic paths on flat surfaces, comparing length and singularity counts.
problem Comparing geometric length and singularity counts on geodesic paths.
method Apply counting limit laws to infinite graphs and then to flat surfaces.
result Statistical comparison of geometric length and singularity counts on geodesic paths.
A web browser should not be only for browsing web pages but also help users to find out their target websites and recommend similar type websites based on their behavior. Throughout this paper, we propose two methods to make a web browser more intelligent about link prediction which works during typing on address-bar a…
The paper develops personalized DAG models for web user behavior.
problem Understanding user behavior transitions between websites with user heterogeneity and network dependency.
method Personalized Binomial DAG models with network-structured covariates, embedding network structure into a dimension-reduced covariate, learning node neighborhoods, and exploring variance-mean relation.
result The proposed algorithm outperforms state-of-the-art competitors in heterogeneous data.
Bootstrap method for Markov chains in reinforcement learning.
problem Distributional consistency in finite controlled Markov chains with unknown control policies.
method Model-based bootstrap with novel LLN and CLT for visitation counts and transition increments.
result Asymptotically valid confidence intervals for value and Q-functions in offline RL. A new approach for deep exploration in sparse reward reinforcement learning.
problem Slow or no learning in reinforcement learning with rare rewards.
method Long-term visitation count planning and decoupling exploration and exploitation.
result Significantly outperforms existing methods in sparse reward environments.
MedGraph learns patient visit embeddings from EMRs, capturing both attributes and temporal sequences.
problem Limited EMR embedding methods fail to capture patient demographics, utilisation, and code descriptions.
method MedGraph constructs an attributed bipartite graph and uses a point process to model temporal sequences.
result MedGraph outperforms state-of-the-art methods in medical risk prediction tasks.
Framework identifies comorbidities for frequent ED and inpatient visits.
problem Reducing resource usage and costs in frequent patients.
method Developed MSAR algorithm to identify comorbidities.
result MSAR identifies conditions most associated with reoccurring ED and inpatient visits.
Curiosity-Critic improves world model training by focusing on cumulative prediction error.
problem Training world models with intrinsic rewards that consider cumulative prediction error.
method Curiosity-Critic uses a surrogate reward based on the difference between current and asymptotic prediction errors, estimated online by a co-trained critic.
result Curiosity-Critic outperforms other methods in training speed and final world model accuracy.
Is it true that patients with similar conditions get similar diagnoses? In this paper we show NLP methods and a unique corpus of documents to validate this claim. We (1) introduce a method for representation of medical visits based on free-text descriptions recorded by doctors, (2) introduce a new method for clustering…
Langevin DQN achieves deep exploration using Gaussian noise.
problem Deep exploration in reinforcement learning.
method Developed Langevin DQN, a variation of DQN with Gaussian noise.
result Langevin DQN achieves deep exploration.
Study automates detection of visitation disruptions in ICU patients.
problem Difficulty in detecting frequent visitation disruptions in ICU patients.
method Used DensePose R-CNN model to count people in video frames, analyzed disruptions and patient outcomes.
result Automated method detects visitation disruptions, impacts on pain and length of stay examined.
EQO uses a simple bonus term for efficient exploration in tabular RL.
problem Achieving minimax optimal reinforcement learning with practical efficiency.
method Exploration via Quasi-Optimism, employing a bonus proportional to state-action visit count.
result Achieves the sharpest known regret bound for tabular RL.
Computational models that forecast the progression of Alzheimer's disease at the patient level are extremely useful tools for identifying high risk cohorts for early intervention and treatment planning. The state-of-the-art work in this area proposes models that forecast by using latent representations extracted from t…
The paper introduces a new intrinsic reward method for exploration in reinforcement learning.
problem Improving exploration in reinforcement learning agents.
method Intrinsic rewards proportional to the entropy of future state-action features.
result The new objective leads to improved visitation of features within individual trajectories.
Real-Time Bidding is nowadays one of the most promising systems in the online advertising ecosystem. In the presented study, the performance of RTB campaigns is improved by optimising the parameters of the users' profiles and the publishers' websites. Most studies about optimising RTB campaigns are focused on the biddi…
Unsupervised model detects healthcare fraud from patient visit data.
problem Detecting fraudulent healthcare bills from patient visit data.
method Uses LSTM and seq2seq models for anomaly detection, normalizes scores with EDF.
result Improves anomaly detection for high class imbalance problems.
The paper analyzes how SGD visits different regions of a non-convex problem's state space.
problem Understanding the long-run distribution of stochastic gradient descent in non-convex problems.
method Large deviations theory and randomly perturbed dynamical systems.
result The long-run distribution of SGD resembles the Boltzmann-Gibbs distribution with temperature equal to the step-size.
VisitHGNN predicts visit probabilities between neighborhoods and POIs using graph neural networks.
problem Estimating visit probabilities between neighborhoods and POIs for urban planning.
method Heterogeneous, relation-specific graph neural network (VisitHGNN) trained on mobility data.
result Strong predictive performance with high fidelity to observed travel behavior.
Understanding how users navigate in a network is of high interest in many applications. We consider a setting where only aggregate node-level traffic is observed and tackle the task of learning edge transition probabilities. We cast it as a preference learning problem, and we study a model where choices follow Luce's a…
New RL approach uses future state and action visitation measures for better exploration.
problem Improving exploration in reinforcement learning.
method Intrinsic reward based on future state and action visitation measures, using contraction operators.
result Policies achieve good state-action space coverage and high performance.
Paper proposes machine learning model for early Alzheimer's diagnosis.
problem Early and accurate diagnosis of Alzheimer's Disease.
method Machine learning models, demographic, biomarker, and cognitive test data.
result 90% accuracy and 87% accuracy in predicting Alzheimer's development.
In recent years, state-of-the-art game-playing agents often involve policies that are trained in self-playing processes where Monte Carlo tree search (MCTS) algorithms and trained policies iteratively improve each other. The strongest results have been obtained when policies are trained to mimic the search behaviour of…
In this paper, we apply neural networks into digital marketing world for the purpose of better targeting the potential customers. To do so, we model the customer online behaviours using dedicated neural network architectures. Starting from user searched keywords in a search engine to the landing page and different foll…
Monte Carlo Tree Search (MCTS) algorithms have achieved great success on many challenging benchmarks (e.g., Computer Go). However, they generally require a large number of rollouts, making their applications costly. Furthermore, it is also extremely challenging to parallelize MCTS due to its inherent sequential nature:…
This article was withdrawn by the arXiv.org administrators since it plagiarizes math.GT/0011056.
Predicting the patient's clinical outcome from the historical electronic medical records (EMR) is a fundamental research problem in medical informatics. Most deep learning-based solutions for EMR analysis concentrate on learning the clinical visit embedding and exploring the relations between visits. Although those wor…
Enhances early risk assessments for pediatric outcomes using contrastive learning.
problem Improving risk assessments in early stages of pediatric development.
method Contrastive multi-modal framework that treats each time window as a distinct modality, training on all available data.
result Consistent improvements in early-stage risk assessments validated on real-world tasks.
New concept: reward hacking, where optimizing a flawed reward function can hurt performance.
problem Optimizing imperfect reward functions leads to poor performance.
method Formal definition and analysis of reward hacking, examining conditions for unhackability.
result Reward functions are usually hackable, making it hard to align AI with human values.
We show how to learn low-dimensional representations (embeddings) of patient visits from the corresponding electronic health record (EHR) where International Classification of Diseases (ICD) diagnosis codes are removed. We expect that these embeddings will be useful for the construction of predictive statistical models…
Unified framework for distributional regret in bandits and reinforcement learning.
problem Characterizing the distribution of regret in multi-armed bandits and reinforcement learning.
method Unified framework with a UCBVI-style algorithm and distributional regret bounds.
result Distributional regret bounds with optimal trade-offs between expected and distributional regret.
Modeling infection hotspots to quantify effects of contact tracing and testing.
problem Capturing the role of infection hotspots in disease transmission.
method Temporal point process modeling framework to represent visits and disease transmission.
result Estimation of transmission rates at sites and households using Bayesian optimization.
Pediatric asthma is the most prevalent chronic childhood illness, afflicting about 6.2 million children in the United States. However, asthma could be better managed by identifying and avoiding triggers, educating about medications and proper disease management strategies. This research utilizes deep learning methodolo…
Study online RL with mismatched dynamics, achieving sublinear regret.
problem Exploration challenges in online RL with mismatched training and deployment dynamics.
method Introduce supremal visitation ratio, propose efficient algorithm with f-divergence. result Achieves sublinear regret in online RMDPs with optimal dependence on supremal visitation ratio and interaction episodes.
Counting tripods on a flat torus using lattice point counting.
problem Counting finite BPS webs in flat torus geometry.
method Lattice point counting techniques in C2. result Asymptotic counting result for tripods on the torus.
Transforming sparse outcomes into dense process rewards for efficient reinforcement learning.
problem Training RL policies to maximize sparse outcomes.
method Incentivizing policy matching state-action visitations of successful episodes.
result Significantly faster RL finetuning performance.
The paper introduces return parity for fairness in MDPs, addressing delayed and adverse effects.
problem Fairness in MDPs for dynamic domains with delayed and adverse effects.
method Proposes return parity, decomposes return disparity, and develops algorithms for state visitation distributional alignment.
result The proposed algorithms can successfully close the disparity gap while maintaining policy performance.
Algorithm optimizes constrained reinforcement learning with dual variables.
problem Minimizing convex functional subject to convex constraint in large state spaces.
method VPDPO algorithm using Lagrangian and Fenchel duality.
result Achieves sublinear regret and constraint violation, globally optimal policy.
Flow Matching for count data improves sample quality and efficiency.
problem Mapping between count distributions across batches or time points in high-dimensional count data.
method count-FM, a flow-matching framework based on a continuous-time birth-death process with local unit jumps.
result count-FM achieves better sample quality than representative baselines while using fewer parameters.
Study how untrained policies explore in RL environments.
problem Challenges in reinforcement learning, especially sparse or adversarial reward structures.
method Theoretical and empirical analysis of untrained deep neural policies in a toy model.
result Untrained policies generate correlated actions and non-trivial state-visitation distributions.
New theorem counts curves on orbifolds.
problem Counting curves on surfaces.
method Applied Mirzakhani's theorem to orbifolds.
result Curve counting theorem extends to orbifolds.
We present here a general framework and a specific algorithm for predicting the destination, route, or more generally a pattern, of an ongoing journey, building on the recent work of [Y. Lassoued, J. Monteil, Y. Gu, G. Russo, R. Shorten, and M. Mevissen, "Hidden Markov model for route and destination prediction," in IE…
Effective modeling of electronic health records presents many challenges as they contain large amounts of irregularity most of which are due to the varying procedures and diagnosis a patient may have. Despite the recent progress in machine learning, unsupervised learning remains largely at open, especially in the healt…
A new method, Count-MORL, improves offline reinforcement learning by using state-action frequency.
problem Improving offline reinforcement learning performance.
method Integrates count-based conservatism into model-based offline reinforcement learning.
result The learned policy is near-optimal and outperforms existing methods.
Proposes a method to reconcile count time series forecasts.
problem No formal framework for probabilistic reconciliation of count time series.
method Generalizes Bayes' rule for reconciling real-valued and count variables.
result Improves forecast accuracy for count variables compared to Gaussian reconciliation.
BRPO optimizes batch RL policies to better exploit state-action differences.
problem Batch RL's conservatism limits exploitation of state-action differences.
method Proposes residual policies and derives BRPO to maximize policy performance.
result BRPO achieves state-of-the-art performance in various tasks.
Graph neural networks struggle with counting certain substructures in graphs.
problem Detecting and counting specific substructures in graphs.
method Study of graph neural networks' ability to count attributed graph substructures.
result Graph neural networks like MPNNs, 2-WL, and 2-IGNs have limitations in counting certain substructures.