The promise of reinforcement learning is to solve complex sequential decision problems autonomously by specifying a high-level reward function only. However, algorithms struggle when, as often the case, simple and intuitive rewards provide sparse deceptive feedback. Avoiding these pitfalls requires thoroughly exploring environment, but creating that can do so remains one central challenges fiel...

Do we desire something because it has value? Or is the value of a thing a consequence of the fact that it is desired? To adopt the former alternative is to involve oneself in the task of providing a theory of value which would make value a property things have as it were prior to their serving as objects of desire. Theories of this sort have been developed inter alia by Meinong, Nicolai Hartman...

We present a variant of Reimer’s proof of the van den Berg-Kesten conjecture.

Death during autoerotic episodes is of special concern to law enforcement officials, the coroner or medical examiner, the family of the decedent, and to society as a whole. As in the probing of any violent demise, accurate preservation of all evidentiary material, complete photographic documentation, reconstruction of the scene, and interviews with family and acquaintances ("psychological autop...

Virtual environments are becoming more and more accepted, and part of the everyday online experience for many users. This offers new potential for both surveillance and data mining. Some of the techniques used are discussed in this paper. However, such activities may in many countries conflict with the legal framework in place, for example with the German Federal Data Protection Act (Datenschut...

0