Hossein Hajimirsadeghi

Multiobjective invasive weed optimization: Application to analysis of Pareto improvement models in electricity markets

This paper presents a proposal for multiobjective Invasive Weed Optimization (IWO) based on nondominated sorting of the solutions. IWO is an ecologically inspired stochastic optimization algorithm which has shown successful results for global optimization. In the present work, performance of the proposed nondominated sorting IWO (NSIWO) algorithm is evaluated through a number of well-known benchmarks for multiobjective optimization. The simulation results of the test problems show that this algorithm is comparable with other multiobjective evolutionary algorithms and is also capable of finding better spread of solutions in some cases. Next, the proposed algorithm is employed to study the Pareto improvement model in two complex electricity markets. First, the Pareto improvement solution set is obtained for a three-player oligopolistic electricity market with a nonlinear demand function. Then, the IEEE 30-bus power system with transmission constraints is considered, and the Pareto improvement solutions are found for the model with deterministic cost functions. In addition, NSIWO algorithm is used to analyze this system with stochastic cost data in a risk management problem which maximizes the expected total profit but minimizes the profit risk in the market.

Multimedia event detection with multimodal feature fusion and temporal concept localization

We present a system for multimedia event detection. The developed system characterizes complex multimedia events based on a large array of multimodal features, and classifies unseen videos by effectively fusing diverse responses. We present three major technical innovations. First, we explore novel visual and audio features across multiple semantic granularities, including building, often in an unsupervised manner, mid-level and high-level features upon low-level features to enable semantic understanding. Second, we show a novel Latent SVM model which learns and localizes discriminative high-level concepts in cluttered video sequences. In addition to improving detection accuracy beyond existing approaches, it enables a unique summary for every retrieval by its use of high-level concepts and temporal evidence localization. The resulting summary provides some transparency into why the system classified the video as it did. Finally, we present novel fusion learning algorithms and our methodology to improve fusion learning under limited training data condition. Thorough evaluation on a large TRECVID MED 2011 dataset showcases the benefits of the presented system.

A hybrid IWO/PSO algorithm for fast and global optimization

This paper presents a hybrid optimization algorithm which originates from Invasive Weed Optimization (IWO) and Particle Swarm Optimization (PSO). Based on the novel and distinct qualifications of IWO and PSO, we introduce IWO/PSO algorithm and try to combine their excellent features in this extended algorithm. The efficiency of this algorithm both in the case of speed of convergence and optimality of the results are compared with IWO, PSO, and some other evolutionary algorithms through a number of common multi-dimensional benchmark functions. Finally, a practical problem consisting design and optimization of an adaptive controller for a surge tank is simulated. The experimental results show that the proposed algorithm can be successfully employed as a fast and global optimization method for a variety of theoretical or practical purposes.

Discrete invasive weed optimization algorithm: application to cooperative multiple task assignment of UAVs

This paper presents a novel discrete population based stochastic optimization algorithm inspired from weed colonization. Its performance in a discrete benchmark, time-cost trade-off (TCT) problem, is evaluated and compared with five other evolutionary algorithms. Also we use our proposed discrete invasive weed optimization (DIWO) algorithm for cooperative multiple task assignment of unmanned aerial vehicles (UAVs) and compare the solutions with those of genetic algorithms (GAs) which have shown satisfactory results in the previous works. UAV task assignment problem is of great interest among researchers and many deterministic and stochastic methods have been devised to come up with the problem. Monte Carlo simulations show successful results that verify better performance of DIWO compared to GAs in both optimality of the solutions and computational time.

Visual recognition by counting instances: A multi-instance cardinality potential kernel

Many visual recognition problems can be approached by counting instances. To determine whether an event is present in a long internet video, one could count how many frames seem to contain the activity. Classifying the activity of a group of people can be done by counting the actions of individual people. Encoding these cardinality relationships can reduce sensitivity to clutter, in the form of irrelevant frames or individuals not involved in a group activity. Learned parameters can encode how many instances tend to occur in a class of interest. To this end, this paper develops a powerful and flexible framework to infer any cardinality relation between latent labels in a multi-instance model. Hard or soft cardinality relations can be encoded to tackle diverse levels of ambiguity. Experiments on tasks such as human activity recognition, video event detection, and video summarization demonstrate the effectiveness of using cardinality relations for improving recognition results.

Cooperative coevolutionary invasive weed optimization and its application to Nash equilibrium search in electricity markets

This paper presents a coevolutionary algorithm named cooperative coevolutionary invasive weed optimization (CCIWO) and investigates its performance for global optimization of functions with numerous local optima and also Nash equilibrium (NE) search for games. Ability of CCIWO for function optimization is tested through a set of common benchmarks of stochastic optimization, and reported results are compared with two other coevolutionary algorithms. In advance, a three-bus transmission-constrained electricity market model is studied, and CCIWO is employed to find NE for this complex system. Experimental results show efficiency of the proposed method to have more accurate solutions.

Discovering human interactions in videos with limited data labeling

We present a novel approach for discovering human interactions in videos. Activity understanding techniques usually require a large number of labeled examples, which are not available in many practical cases. Here, we focus on recovering semantically meaningful clusters of human-human and human-object interaction in an unsupervised fashion. A new iterative solution is introduced based on Maximum Margin Clustering (MMC), which also accepts user feedback to refine clusters. This is achieved by formulating the whole process as a unified constrained latent max-margin clustering problem. Extensive experiments have been carried out over three challenging datasets, Collective Activity, VIRAT, and UT-interaction. Empirical results demonstrate that the proposed algorithm can efficiently discover perfect semantic clusters of human interactions with only a small amount of labeling effort.

Conceptual Imitation Learning in a Human-Robot Interaction Paradigm

In general, imitation is imprecisely used to address different levels of social learning from high-level knowledge transfer to low-level regeneration of motor commands. However, true imitation is based on abstraction and conceptualization. This article presents a model for conceptual imitation through interaction with the teacher to abstract spatio-temporal demonstrations based on their functional meaning. Abstraction, concept acquisition, and self-organization of proto-symbols are performed through an incremental and gradual learning algorithm. In this algorithm, Hidden Markov Models (HMMs) are used to abstract perceptually similar demonstrations. However, abstract (relational) concepts emerge as a collection of HMMs irregularly scattered in the perceptual space but showing the same functionality. Performance of the proposed algorithm is evaluated in two experimental scenarios. The first one is a human-robot interaction task of imitating signs produced by hand movements. The second one is a simulated interactive task of imitating whole body motion patterns of a humanoid model. Experimental results show efficiency of our model for concept extraction, proto-symbol emergence, motion pattern recognition, prediction, and generation.

Learning Ensembles of Potential Functions for Structured Prediction with Latent Variables

Many visual recognition tasks involve modeling variables which are structurally related. Hidden conditional random fields (HCRFs) are a powerful class of models for encoding structure in weakly supervised training examples. This paper presents HCRF-Boost, a novel and general framework for learning HCRFs in functional space. An algorithm is proposed to learn the potential functions of an HCRF as a combination of abstract nonlinear feature functions, expressed by regression models. Consequently, the resulting latent structured model is not restricted to traditional log-linear potential functions or any explicit parameterization. Further, functional optimization helps to avoid direct interactions with the possibly large parameter space of nonlinear models and improves efficiency. As a result, a complex and flexible ensemble method is achieved for structured prediction which can be successfully used in a variety of applications. We validate the effectiveness of this method on tasks such as group activity recognition, human action recognition, and multi-instance learning of video events.

Computer Vision Techniques to Collect Helmet-Wearing Data on Cyclists

Several studies have shown that cyclists can reduce the risk of severe head injuries by wearing a helmet. A system is proposed to collect cyclist helmet usage data automatically from video footage. Computer vision techniques are used to track the moving objects and then to analyze the object trajectories and speed profiles to identify cyclists. Image features are extracted from a region around the cyclists head. Support vector machines determine whether the cyclist is wearing a helmet. The system can be approximately 90% accurate in cyclist classification when provided with accurate tracks of the cyclists head. Even for situations in which obtaining video to track a cyclist is challenging, the proposed method provides an effective retrieval system, potentially reducing the number of video records that must be analyzed manually to find instances of cyclists not wearing helmets.