Arrow Research search

Author name cluster

Eran Halperin

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

14 papers
2 author rows

Possible papers

14

TMLR Journal 2025 Journal Article

Guided Discrete Diffusion for Electronic Health Record Generation

  • Jun Han
  • Zixiang Chen
  • Yongqian Li
  • Yiwen Kou
  • Eran Halperin
  • Robert E. Tillman
  • Quanquan Gu

Electronic health records (EHRs) are a pivotal data source that enables numerous applications in computational medicine, e.g., disease progression prediction, clinical trial design, and health economics and outcomes research. Despite wide usability, their sensitive nature raises privacy and confidentially concerns, which limit potential use cases. To tackle these challenges, we explore the use of generative models to synthesize artificial, yet realistic EHRs. While diffusion-based methods have recently demonstrated state-of-the-art performance in generating other data modalities and overcome the training instability and mode collapse issues that plague previous GAN-based approaches, their applications in EHR generation remain underexplored. The discrete nature of tabular medical code data in EHRs poses challenges for high-quality data generation, especially for continuous diffusion models. To this end, we introduce a novel tabular EHR generation method, EHR-D3PM, which enables both unconditional and conditional generation using the discrete diffusion model. Our experiments demonstrate that EHR-D3PM significantly outperforms existing generative baselines on comprehensive fidelity and utility metrics while maintaining less attribute and membership vulnerability risks. Furthermore, we show EHR-D3PM is effective as a data augmentation method and enhances performance on downstream tasks when combined with real data.

PDF Details

AAAI Conference 2025 Conference Paper

Memorize and Rank: Elevating Large Language Models for Clinical Diagnosis Prediction

  • Mingyu Derek Ma
  • Xiaoxuan Wang
  • Yijia Xiao
  • Anthony Cuturrufo
  • Vijay S Nori
  • Eran Halperin
  • Wei Wang

Clinical diagnosis prediction models, when provided with a patient's medical history, aim to detect potential diseases early, facilitating timely intervention and improving prognostic outcomes. However, the inherent scarcity of patient data and large disease candidate space often pose challenges in developing satisfactory models for this intricate task. The exploration of leveraging Large Language Models (LLMs) for encapsulating clinical decision processes has been limited. We introduce MERA, a clinical diagnosis prediction model that bridges pertaining natural language knowledge with medical practice. We apply hierarchical contrastive learning on a disease candidate ranking list to alleviate the large decision space issue. With concept memorization through fine-tuning, we bridge the natural language clinical knowledge with medical codes. Experimental results on MIMIC-III and IV datasets show that MERA achieves the state-of-the-art diagnosis prediction performance and dramatically elevates the diagnosis prediction capabilities of generative LMs.

PDF Details DOI

TCS Journal 2005 Journal Article

Performance aspects of distributed caches using TTL-based consistency

  • Edith Cohen
  • Eran Halperin
  • Haim Kaplan

The web is the largest distributed database deploying time-to-live-based weak consistency. Each object has a lifetime-duration assigned to it by its origin server. A copy of the object fetched from its origin server is received with maximum time-to-live (TTL) that equals its lifetime duration. In contrast a copy obtained through a cache have shorter TTL since the age (elapsed time since fetched from the origin) is deducted from its lifetime duration. A request served by a cache constitutes a hit if the cache has a fresh copy of the object. Otherwise, the request is considered a miss and is propagated to another server. It is evident that the number of cache misses depends on the age of the copies the cache receives. Thus, a cache that sends requests to another cache would suffer more misses than a cache that sends requests directly to an authoritative server. In this paper, we model and analyze the effect of age on the performance of various cache configurations. We consider a low-level cache that fetches objects either from their origin servers or from other caches and analyze its miss-rate as function of its fetching policy. We distinguish between three basic fetching policies, namely, fetching always from the origin, fetching always from the same high-level cache, and fetching from a “random” high-level cache. We explore the relationships between these policies in terms of the miss-rate achieved by the low-level cache, both on worst-case sequences, and on sequences generated using particular probability distributions. Guided by web caching practice, we consider two variations of the basic policies. In the first variation the high-level cache uses pre-term refreshes to keep a copy with lower age. In the second variation the low-level cache uses extended lifetime duration. We analyze how these variations affect the miss-rates. Our theoretical results help to understand how age may affect the miss-rate, and imply guidelines for improving performance of web caches.

Details DOI

TCS Journal 2005 Journal Article

The minimum-entropy set cover problem

  • Eran Halperin
  • Richard M. Karp

We consider the minimum entropy principle for learning data generated by a random source and observed with random noise. In our setting we have a sequence of observations of objects drawn uniformly at random from a population. Each object in the population belongs to one class. We perform an observation for each object which determines that it belongs to one of a given set of classes. Given these observations, we are interested in assigning the most likely class to each of the objects. This scenario is a very natural one that appears in many real life situations. We show that under reasonable assumptions finding the most likely assignment is equivalent to the following variant of the set cover problem. Given a universe U and a collection S = ( S 1, …, S t ) of subsets of U, we wish to find an assignment f: U → S such that u ∈ f ( u ) and the entropy of the distribution defined by the values | f - 1 ( S i ) | is minimized. We show that this problem is NP-hard and that the greedy algorithm for set cover s with an additive constant error with respect to the optimal cover. This sheds a new light on the behavior of the greedy set cover algorithm. We further enhance the greedy algorithm and show that the problem admits a polynomial time approximation scheme (PTAS). Finally, we demonstrate how this model and the greedy algorithm can be useful in real life scenarios, and in particular, in problems arising naturally in computational biology.

Details DOI

STOC Conference 2003 Conference Paper

A stochastic process on the hypercube with applications to peer-to-peer networks

  • Micah Adler
  • Eran Halperin
  • Richard M. Karp
  • Vijay V. Vazirani

Consider the following stochastic process executed on a graph G=(V,E) whose nodes are initially uncovered. In each step, pick a node at random and if it is uncovered, cover it. Otherwise, if it has an uncovered neighbor, cover a random uncovered neighbor. Else, do nothing. This can be viewed as a structured coupon collector process. We show that for a large family of graphs, O(n) steps suffice to cover all nodes of the graph with high probability, where n is the number of vertices. Among these graphs are d -regular graphs with d =Ω(log n log log n) , random d -regular graphs with d =Ω(log n) and the k -dimensional hypercube where n=2 k .This process arises naturally in answering a question on load balancing in peer-to-peer networks. We consider a distributed hash table in which keys are partitioned across a set of processors, and we assume that the number of processors grows dynamically, starting with a single processor. If at some stage there are n processors, the number of queries required to find a key is log 2 n+O(1) , the number of pointers maintained by each processor is log 2 n+O(1) , and moreover the worst ratio between the loads of processors is O(1) , with high probability. To the best of our knowledge, this is the first analysis of a distributed hash table that achieves asymptotically optimal load balance, while still requiring only O(log n) pointers per processor and O(log n) queries for locating a key; previous methods required Ω(log 2 n) pointers per processor and Ω(log n) queries for locating a key.

Details

STOC Conference 2003 Conference Paper

Polylogarithmic inapproximability

  • Eran Halperin
  • Robert Krauthgamer

We provide the first hardness result of a polylogarithmic approximation ratio for a natural NP-hard optimization problem. We show that for every fixed ε>0 , the GROUP-STEINER-TREE problem admits no efficient log 2-ε k approximation, where k denotes the number of groups (or, alternatively, the input size), unless NP has quasi polynomial Las-Vegas algorithms. This hardness result holds even for input graphs which are Hierarchically Well-Separated Trees , introduced by Bartal [FOCS, 1996]. For these trees (and also for general trees), our bound is nearly tight with the log-squared approximation currently known. Our results imply that for every fixed ε>0 , the DIRECTED-STEINER TREE problem admits no log 2-ε n --approximation, where n is the number of vertices in the graph, under the same complexity assumption.

Details
v2026.09.13