It is widely accepted that the use of more compact representations than lookup tables is crucial to scaling reinforcement learning (RL) algorithms to real-world problems. Unfortunately almost all of the theory of reinforcement learning assumes lookup table representations. In this paper we address the pressing issue of combining function approximation and RL, and present 1) a function approx-im...

Since imidazole buffers have been used in platelet function tests and the compound has been reported to alter several biochemical activities of platelets, it seemed important to determine whether imidazole influenced platelet aggregation. ADP-induced, collagen-induced, and norepinephrine-induced platelet aggregations were tested in platelet-rich plasma by turbidimetric techniques. Glass beadind...

It is widely accepted that the use of more compact representations than lookup tables is crucial to scaling reinforcement learning RL algorithms to real world problems Unfortunately almost all of the theory of reinforcement learning assumes lookup table representa tions In this paper we address the pressing issue of combining function approximation and RL and present a function approx imator ba...

We propose a novel query-dependent feature aggregation (QDFA) method for medical image retrieval. The QDFA method can learn an optimal feature aggregation function for a multi-example query, which takes into account multiple features and multiple examples with different importance. The experiments demonstrate that the QDFA method outperforms three other feature aggregation methods. key words: C...

We present an algorithm for implementing aggregation/broadcast over existing Distributed Hash Tables (DHTs). Our algorithm builds an aggregation tree in a bottom-up fashion by mapping nodes to their parents in the aggregation tree with pre-determined function and maintains it using soft-state. We believe that the aggregation/broadcast function is a basic service that should be layered over exis...

This paper investigates the use of individual cross section data to describe macro functions. Necessary and sufficient conditions (denoted AS) are found for OLS slope coefficients from a cross section to consistently estimate the first derivatives of the macro function. AS embodies both sets of aggregation assumptions known; linear aggregation and sufficient statistics , and thus represents gen...

The SDSN-Aggregation: The SDSN aggregation method uses a combination of methods ”Brooks-Iyengar algorithm” and ”Witness based approach for data fusion assurance” to achieve the aggregation. The aggregation method should work for multiple data types and for different densities of the network. The SDSN-aggregation is event-based, it is triggered on an aggregation request by another node or when n...