Reinforcement learning has been applied to many interesting problems such as the famous TD-gammon [1] and the inverted helicopter flight [2]. However little effort has been put into developing methods to learn policies for complex persistent tasks and tasks that are time-sensitive. In this paper we take a step towards solving this problem by using signal temporal logic (STL) as task specificati...

The C++ Standard Template Library (STL) provides generic, high-performance algorithms and data types. STL also defines performance guarantees, but it is left as a non-trivial task to the library user to choose appropriate algorithms and data types for a particular problem. In the present paper, we describe an analysis to aid library users in selecting types for use in a generic programs. The an...

We apply salient feature detection and tracking in videos to simulate fixations and smooth pursuit in human vision. With tracked sequences as input, a hierarchical network of modules learns invariant features using a temporal slowness constraint. The network encodes invariance which are increasingly complex with hierarchy. Although learned from videos, our features are spatial instead of spatia...

This paper describes experiments done to demonstrate the effectiveness of Multi-task Learning (MTL) for software agents. The experiments were carried out on an agent for processing electronic mail (email). MTL is shown to slightly improve the learning rate in this domain over Single-task Learning (STL) in a k-nearest-neighbor implementation. We then introduce a new method of lazy learning we ca...

We define the target, mechanism, and structural basis of inhibition of bacterial RNA polymerase (RNAP) by the tetramic acid antibiotic streptolydigin (Stl). Stl binds to a site adjacent to but not overlapping the RNAP active center and stabilizes an RNAP-active-center conformational state with a straight-bridge helix. The results provide direct support for the proposals that alternative straigh...

In this paper we describe AMT, a tool for monitoring temporal properties of continuous signals. We first introduce STL/PSL, a specification formalism based on the industrial standard language PSL and the real-time temporal logic MITL, extended with constructs that allow describing behaviors of real-valued variables. The tool automatically builds property observers from an STL/PSL specification ...

Reconstructing a large skull defect is challenge, as it normally involves the use of sophisticated proprietary image processing and expensive CAD software. As an alternative, open-source software can be used for this purpose. This study aimed to compare 3D cranial implants reconstructed from computed tomography (CT) images using MITK with commercial 3-matic ten decompressive craniectomy patient...

This paper presents provable work-optimal parallelizations of STL (Standard Template Library) algorithms based on the workstealing technique. Unlike previous approaches where a deque for each processor is typically used to locally store ready tasks and where a processor that runs out of work steals a ready task from the deque of a randomly selected processor, the current paper instead presents ...

We extend Signal Temporal Logic (STL) to enable the specification of importance and priorities. The extension, called Weighted STL (wSTL), has same qualitative (Boolean) semantics as STL, but additionally defines weights associated with Boolean temporal operators that modulate its quantitative (robustness). show robustness wSTL can be defined weighted generalizations all known compatible functi...