Browsing RPI Theses Open Access by Author "Magdon-Ismail, Malik"
Now showing items 1-20 of 27
-
A step towards the practical evaluation of wildlife identification algorithms
Mankowski, Alexander R. (Rensselaer Polytechnic Institute, Troy, NY, 2022-08)Computer vision approaches have shown to be an effective tool for wildlife identification across a wide range of species. As the field grows and the amount of available data grows with it, deep learning based approaches ... -
Active learning of Gaussian mixture models using direct estimation of error reduction
Gaston, Jeffry (Rensselaer Polytechnic Institute, Troy, NY, 2012-05) -
Approximating covariance matrices using low rank perturbations with applications to accent identification and social network clustering
Purnell, Jonathan (Rensselaer Polytechnic Institute, Troy, NY, 2010-08) -
Combinatorial neural network training algorithm for neuromorphic computing
Date, Prasanna (Rensselaer Polytechnic Institute, Troy, NY, 2019-08)In Computer Science, we have realized that the end of Moore’s Law is just around the corner, and it would not be possible to sustain the exponential increase in computation speed on conventional compute platforms like CPUs ... -
Community evolution in temporal networks
Thompson, James (Rensselaer Polytechnic Institute, Troy, NY, 2015-05)The purpose of this research is to study the structure of social networks with an added temporal element. Specifically, we examine dynamic community behavior within social networks. We base our experiments on a simple ... -
Data analytics of time-series for complex (biological) systems
Dhulekar, Nimit (Rensselaer Polytechnic Institute, Troy, NY, 2015-05)Complex time-series systems such as biological networks have been studied for many years using conventional molecular and cellular techniques. However, the multiscale nature of these networks make these techniques limited ... -
Deciding who, what, why, and how : aggregating preferences over agents, alternatives, axioms, and rules
Abramowitz, Ben (Rensselaer Polytechnic Institute, Troy, NY, 2021-08)Collective decision making requires dealing with competing preferences. One probably needs little convincing that such problems can be complex and difficult to solve. There are many types of preferences, many different ... -
Empirical analysis of optimization algorithms for portfolio allocation
Bolin, Andrew (Rensselaer Polytechnic Institute, Troy, NY, 2013-05)Portfolio optimization algorithms were tested using historical S&P100 data. A traditional Mean-Var algorithm is tested as well as two alternative risk methods. The alternative risk methods used the maximum drawdown (MDD) ... -
Empirical analysis of sparse principal component analysis
Mastylo, Damian Z. (Rensselaer Polytechnic Institute, Troy, NY, 2016-05)Many optimizations exist via tweaking the sparsity factor, the number of left singular vectors used, or the column subset selection method. Many combinations of these approaches are examined, and their efficacy are reported ... -
Exact and approximate equilibria for network formation and cut games
Caskurlu, Bugra (Rensselaer Polytechnic Institute, Troy, NY, 2010-08) -
High quality stable solutions in social and communication networks
Bhardwaj, Onkar (Rensselaer Polytechnic Institute, Troy, NY, 2015-05)Network Formation Games is a rich category of games which studies how a network emerges as an outcome of strategic interaction of self-interested nodes (agents) pursuing their own objectives. The conflicts among node ... -
Matching, social welfare and ordinal approximation
Zhu, Wennan (Rensselaer Polytechnic Institute, Troy, NY, 2020-05)Many important problems involve agents with preferences for different outcomes. Such settings include, for example, social choice and matching problems. Although the quality of an outcome to an agent may be measured by a ... -
Mining local and global patterns for complex data classification
Li, Geng (Rensselaer Polytechnic Institute, Troy, NY, 2013-08)In the second part of this thesis, we study the challenging problem of graph classification. With the proliferation of graph data, there has been a lot of interest in recent years to develop effective methods for classifying ... -
MLCryptoBox : cryptographic toolbox for distributed machine learning
Ishaq, Muhammad (Rensselaer Polytechnic Institute, Troy, NY, 2018-05)We propose a hybrid design using multiparty computation (MPC) to achieve both accuracy and speed. An ML application should compute only those primitives cryptographically securely which operate on sensitive input. Towards ... -
Modeling human behavior in the context of social media during extreme events caused by natural hazards
Tyshchuk, Yulia (Rensselaer Polytechnic Institute, Troy, NY, 2015-05)The results of this research provide a means for behavioral interventions to facilitate the diffusion of critical warning information on social media, diffusion of confirmations, and facilitation of the evacuation. ... -
Non-discriminative algorithmic pricing: decentralized resource allocation in markets
Sekar, Shreyas (Rensselaer Polytechnic Institute, Troy, NY, 2017-05)2) Pricing under uncertainty: Even when the seller only has access to a distribution over (submodular) buyer valuations, and items have a non-linear production cost, we show that it is possible to efficiently compute prices ... -
Optimizing diffusion and pricing in networks of independent agents
Hate, Ameya (Rensselaer Polytechnic Institute, Troy, NY, 2012-05) -
Predicting change points in multivariate time series data
Khan, Haidar (Rensselaer Polytechnic Institute, Troy, NY, 2019-05)Time series data produced by a dynamic system may contain multiple states characterized by distinct patterns. Predicting the phase transitions between them is critical for many important applications including automatic ... -
Random projections for support vector machines
Paul, Saurabh (Rensselaer Polytechnic Institute, Troy, NY, 2012-12)Let X ∈ R n×d be a data matrix of rank ρ, representing n points in R d . The linear support vector machine constructs a hyperplane separator that maximizes the 1-norm soft margin. We develop a new oblivious dimension ... -
Reducing image classification error and quantifying the contributions of components in a learning pipeline
Chowdhury, Aritra (Rensselaer Polytechnic Institute, Troy, NY, 2018-08)In the second part of the Thesis, we try to understand the error contributions from different components of image classification pipelines. In the first work of this part, we are trying to quantify the quality of the ...