Search Constraints
Filtering by:
Date Created
2014
Remove constraint Date Created: 2014
Language
English
Remove constraint Language: English
Resource Type
Conference Proceeding
Remove constraint Resource Type: Conference Proceeding
1 - 5 of 5
Number of results to display per page
Search Results
-
- Resource Type:
- Conference Proceeding
- Creator:
- Bucking, Scott, Zmeureanu, Radu, and Athienitis, Andreas
- Abstract:
- This paper presents a multi-objective redesign case study of an archetype solar house based on a near net zero energy (NZE) demonstration home located in Eastman, Quebec. Using optimization techniques, pathways are identified from the original design to both cost and energy optimal designs. An evolutionary algorithm is used to optimize trade-offs between passive solar gains and active solar generation, using two objective functions: net-energy consumption and life-cycle cost over a thirty-year life cycle. In addition, this paper explores different pathways to net zero energy based on economic incentives, such as feed-in tariffs for on-site electricity production from renewables. The main objective is to identify pathways to net zero energy that will facilitate the future systematic design of similar homes based on the concept of the archetype that combines passive solar design; energy-efficiency measures, including a geothermal heat pump; and a building-integrated photovoltaic system. Results from this paper can be utilized as follows: (1) systematic design improvements and applications of lessons learned from a proven NZE home design concept, (2) use of a methodology to understand pathways to cost and energy optimal building designs, and (3) to aid in policy development on economic incentives that can positively influence optimized home design.
- Date Created:
- 2014-01-01
-
- Resource Type:
- Conference Proceeding
- Creator:
- Guo, Yuhong and Li, Xin
- Abstract:
- Semantic scene classification is a challenging problem in computer vision. In this paper, we present a novel multi-level active learning approach to reduce the human annotation effort for training robust scene classification models. Different from most existing active learning methods that can only query labels for selected instances at the target categorization level, i.e., the scene class level, our approach establishes a semantic framework that predicts scene labels based on a latent object-based semantic representation of images, and is capable to query labels at two different levels, the target scene class level (abstractive high level) and the latent object class level (semantic middle level). Specifically, we develop an adaptive active learning strategy to perform multi-level label query, which maintains the default label query at the target scene class level, but switches to the latent object class level whenever an "unexpected" target class label is returned by the labeler. We conduct experiments on two standard scene classification datasets to investigate the efficacy of the proposed approach. Our empirical results show the proposed adaptive multi-level active learning approach can outperform both baseline active learning methods and a state-of-the-art multi-level active learning method.
- Date Created:
- 2014-01-01
-
- Resource Type:
- Conference Proceeding
- Creator:
- Maheshwari, Anil, Nandy, Ayan, Smid, Michiel, and Das, Sandip
- Abstract:
- Consider a line segment R consisting of n facilities. Each facility is a point on R and it needs to be assigned exactly one of the colors from a given palette of c colors. At an instant of time only the facilities of one particular color are 'active' and all other facilities are 'dormant'. For the set of facilities of a particular color, we compute the one dimensional Voronoi diagram, and find the cell, i.e, a segment of maximum length. The users are assumed to be uniformly distributed over R and they travel to the nearest among the facilities of that particular color that is active. Our objective is to assign colors to the facilities in such a way that the length of the longest cell is minimized. We solve this optimization problem for various values of n and c. We propose an optimal coloring scheme for the number of facilities n being a multiple of c as well as for the general case where n is not a multiple of c. When n is a multiple of c, we compute an optimal scheme in Θ(n) time. For the general case, we propose a coloring scheme that returns the optimal in O(n2logn) time.
- Date Created:
- 2014-01-01
-
- Resource Type:
- Conference Proceeding
- Creator:
- Guo, Yuhong and Li, Xin
- Abstract:
- Multi-label classification is a central problem in many application domains. In this paper, we present a novel supervised bi-directional model that learns a low-dimensional mid-level representation for multi-label classification. Unlike traditional multi-label learning methods which identify intermediate representations from either the input space or the output space but not both, the mid-level representation in our model has two complementary parts that capture intrinsic information of the input data and the output labels respectively under the autoencoder principle while augmenting each other for the target output label prediction. The resulting optimization problem can be solved efficiently using an iterative procedure with alternating steps, while closed-form solutions exist for one major step. Our experiments conducted on a variety of multi-label data sets demonstrate the efficacy of the proposed bi-directional representation learning model for multi-label classification.
- Date Created:
- 2014-01-01
-
- Resource Type:
- Conference Proceeding
- Creator:
- Bose, Prosenjit and Van Renssen, André
- Abstract:
- We present tight upper and lower bounds on the spanning ratio of a large family of constrained θ-graphs. We show that constrained θ-graphs with 4k2 (k≥ 1 and integer) cones have a tight spanning ratio of 1+2 sin(θ/2), where θ is 2 π/ (4k+2). We also present improved upper bounds on the spanning ratio of the other families of constrained θ-graphs.
- Date Created:
- 2014-01-01