I suggest we can apply AI in different sector, that be can economics, computer science, physics ......etc. In my opinion i think you should be first to fix your objects to attained in AI and finding the problems in what your chose. after that you can find a lot of information for implementation, example: Transfer learning VGG-16, VGG-19, ResNet-50, ResNet-101.....) or Deep learning by architectures: NN , CNN, RNN.....) implemented in famous framework: TensorFlow and keras or Pytorch and Caff.

I hope That be Claire for you.

I have worked with recent Swarm algorithms for handling different projects such as :

1- Feature selection for handling large scale data sets (Big data)

2- Building clustering algorithms for social media analysis

3- Recommender system for Facebook branding pages

4- Prediction apps for medical such as seminal quality, for forest fires disaster prediction , Wind speed forecasting for renewable energy prediction

5- and more..

Please check my profile topics, algorithms, projects based on swarm algorithms

I think this by you what name you are giving to your proposed algorithm? To me all Mathematical/Statistical Models/Algorithms are nature inspired algorithms and you can apply it to almost all Artificial Intelligence related applications/tasks.

1) nature inspired optimisation for feature selection

2) nature inspired optimisation for explainable Artificial Intelligence (XAI)

3) deep learning training with global optimisation algorithms

4) nature inspired fitness function for curiosity based AI

Please look at the class

Artificial intelligence

by

Professor Winston

at

MIT

on

YouTube

There are some nature-inspired optimization algorithms.

For example, ant colony: https://en.wikipedia.org/wiki/Ant_colony_optimization_algorithms

You can use hybrid of machine learning models with nature inspired algorithms for engineering problems such as wind modeling, river flow modeling.

you can go for different domain problems such as domain-specific chat-bot systems (NLP), network traffic speed prediction (transportation systems) as well as E-healthcare systems (AI+Image processing) for breast cancer, cervical cancer, and oral cancer, etc.

Dear Ankita Jiyani,

Mainly When you are using different nature inspired algorithms with ML techniques, generally they are used to optimize the weights of ML models. There are a wide range of applications for it like 1) Software effort estimation 2) Forecasting on different time series data, 3) NLP, 4) Detection of different diseases using Image processing 5) In clustering 6) Robot navigation and path planning 7) Water resource estimation etc..

Hope it may help you in your research.

Thanks

I suggest we can apply AI in different sector, that be can economics, computer science, physics ......etc. In my opinion i think you should be first to fix your objects to attained in AI and finding the problems in what your chose. after that you can find a lot of information for implementation, example: Transfer learning VGG-16, VGG-19, ResNet-50, ResNet-101.....) or Deep learning by architectures: NN , CNN, RNN.....) implemented in famous framework: TensorFlow and keras or Pytorch and Caff.

I hope That be Claire for you.

Solving optimization issues with AI & Nature-inspired algorithms.

Your topic of choice actually depends on application domain. My advice is to focus on improving on any latest optimization algorithms such as Elephant herd optimization (EHO), Grey Wolf Optimization........

Bayesian Deep learning with uncertainty estimation will be a good topic.

Currently, deep learning only generate deterministic results. By also adding uncertainty measure, this will enrich the final results for the decisions.