Goto

Collaborating Authors

Instance-Based Relevance Feedback for Image Retrieval

Neural Information Processing Systems

High retrieval precision in content-based image retrieval can be attained by adopting relevance feedback mechanisms. These mechanisms require that the user judges the quality of the results of the query by marking all the retrieved images as being either relevant or not. Then, the search engine exploits this information to adapt the search to better meet user's needs. At present, the vast majority of proposed relevance feedback mechanisms are formulated in terms of search model that has to be optimized. Such an optimization involves the modification of some search parameters so that the nearest neighbor of the query vector contains the largest number of relevant images.


RT @elkefrank: Three Stages of #AI via @gp_pulipaka #BigData #MachineLearning …

#artificialintelligence

Any price and availability information displayed on [relevant Amazon Site(s), as applicable] at the time of purchase will apply to the purchase of this product.) Any price and availability information displayed on [relevant Amazon Site(s), as applicable] at the time of purchase will apply to the purchase of this product.) Any price and availability information displayed on [relevant Amazon Site(s), as applicable] at the time of purchase will apply to the purchase of this product.) Any price and availability information displayed on [relevant Amazon Site(s), as applicable] at the time of purchase will apply to the purchase of this product.) Any price and availability information displayed on [relevant Amazon Site(s), as applicable] at the time of purchase will apply to the purchase of this product.)


Random Subspace with Trees for Feature Selection Under Memory Constraints

arXiv.org Machine Learning

Dealing with datasets of very high dimension is a major challenge in machine learning. In this paper, we consider the problem of feature selection in applications where the memory is not large enough to contain all features. In this setting, we propose a novel tree-based feature selection approach that builds a sequence of randomized trees on small subsamples of variables mixing both variables already identified as relevant by previous models and variables randomly selected among the other variables. As our main contribution, we provide an in-depth theoretical analysis of this method in infinite sample setting. In particular, we study its soundness with respect to common definitions of feature relevance and its convergence speed under various variable dependance scenarios. We also provide some preliminary empirical results highlighting the potential of the approach.



Rethinking Artificial Intelligence

AITopics Original Links

Intelligent systems for Human Computer Interaction (HCI) will enter into the environment of knowledge workers as capable assistants who respond to problems both by presenting relevant information and more significantly by finding other members of the organization who possess true expertise relevant to the problem at hand. Intelligent HCI systems will not only establish, but also support the interactions between the collaborating problem solvers; in particular, they will have models of interaction styles relevant to different stages of a project and will help at each stage to facilitate interactions with the greatest likelihood of benefit. Intelligent HCI systems will lead to a highly fluid organization in which teams can be established and disbanded as needed. Valuable expertise will be brought to bear on a problem as needed but only as long as needed. Intelligent systems for Human Computer Interaction (HCI) will enter into the environment of knowledge workers as capable assistants who respond to problems both by presenting relevant information and more significantly by finding other members of the organization who possess true expertise relevant to the problem at hand.