Vivek Kumar Sharma

Vivek Sharma, Praneeth Vepakomma, Tristan Swedish et al.

Recently, there has been the development of Split Learning, a framework for distributed computation where model components are split between the client and server (Vepakomma et al., 2018b). As Split Learning scales to include many different model components, there needs to be a method of matching client-side model components with the best server-side model components. A solution to this problem was introduced in the ExpertMatcher (Sharma et al., 2019) framework, which uses autoencoders to match raw data to models. In this work, we propose an extension of ExpertMatcher, where matching can be performed without the need to share the client's raw data representation. The technique is applicable to situations where there are local clients and centralized expert ML models, but the sharing of raw data is constrained.

Vivek Sharma, Praneeth Vepakomma, Tristan Swedish et al.

In this work we introduce ExpertMatcher, a method for automating deep learning model selection using autoencoders. Specifically, we are interested in performing inference on data sources that are distributed across many clients using pretrained expert ML networks on a centralized server. The ExpertMatcher assigns the most relevant model(s) in the central server given the client's data representation. This allows resource-constrained clients in developing countries to utilize the most relevant ML models for their given task without having to evaluate the performance of each ML model. The method is generic and can be beneficial in any setup where there are local clients and numerous centralized expert ML models.

Sandeep Kumar, Vivek Kumar Sharma, Rajani Kumari

Artificial Bee Colony (ABC) optimization algorithm is one of the recent population based probabilistic approach developed for global optimization. ABC is simple and has been showed significant improvement over other Nature Inspired Algorithms (NIAs) when tested over some standard benchmark functions and for some complex real world optimization problems. Memetic Algorithms also become one of the key methodologies to solve the very large and complex real-world optimization problems. The solution search equation of Memetic ABC is based on Golden Section Search and an arbitrary value which tries to balance exploration and exploitation of search space. But still there are some chances to skip the exact solution due to its step size. In order to balance between diversification and intensification capability of the Memetic ABC, it is randomized the step size in Memetic ABC. The proposed algorithm is named as Randomized Memetic ABC (RMABC). In RMABC, new solutions are generated nearby the best so far solution and it helps to increase the exploitation capability of Memetic ABC. The experiments on some test problems of different complexities and one well known engineering optimization application show that the proposed algorithm outperforms over Memetic ABC (MeABC) and some other variant of ABC algorithm(like Gbest guided ABC (GABC),Hooke Jeeves ABC (HJABC), Best-So-Far ABC (BSFABC) and Modified ABC (MABC) in case of almost all the problems.

Sandeep Kumar, Vivek Kumar Sharma, Rajani Kumari

Artificial bee colony (ABC) algorithm has proved its importance in solving a number of problems including engineering optimization problems. ABC algorithm is one of the most popular and youngest member of the family of population based nature inspired meta-heuristic swarm intelligence method. ABC has been proved its superiority over some other Nature Inspired Algorithms (NIA) when applied for both benchmark functions and real world problems. The performance of search process of ABC depends on a random value which tries to balance exploration and exploitation phase. In order to increase the performance it is required to balance the exploration of search space and exploitation of optimal solution of the ABC. This paper outlines a new hybrid of ABC algorithm with Genetic Algorithm. The proposed method integrates crossover operation from Genetic Algorithm (GA) with original ABC algorithm. The proposed method is named as Crossover based ABC (CbABC). The CbABC strengthens the exploitation phase of ABC as crossover enhances exploration of search space. The CbABC tested over four standard benchmark functions and a popular continuous optimization problem.