Research projects in which I worked and I am working on.
The components of a machine learning solution
- Data Generation: Every machine learning application lives off data. That data has to come from somewhere.
- Data Collection: Data is only useful if it’s accessible, so it needs to be stored – ideally in a consistent structure and conveniently in one place.
- Feature Engineering Pipeline: We have to select, transform, combine, and otherwise prepare our data so the algorithm can find useful patterns.
- Training: We apply algorithms, and they learn patterns from the data. Then they use these patterns to perform particular tasks.
- Evaluation: We need to carefully test how well our algorithm performs on data it hasn’t seen before (during training).
- Task Orchestration: Feature engineering, training, and prediction all need to be scheduled on our compute infrastructure (such as AWS or Azure) – usually with non-trivial interdependence.
- Prediction: We use the model we’ve trained to perform new tasks and solve new problems.
- Infrastructure: Even in the age of the cloud, the solution has to live and be served somewhere. This will require setup and maintenance.
- Authentication: This keeps our models secure and makes sure only those who have permission can use them.
- Interaction: We need some way to interact with our model and give it problems to solve. Usually this takes the form of an API.
- Monitoring: We need to regularly check our model’s performance. This usually involves periodically generating a report or showing performance history in a dashboard.
FRIDA (inFRared Imager and Dissector for Adaptive optics) will be an integral field spectrograph (near infrared) with imaging capability for use with the adaptive optics system of the Gran Telescopio CANARIAS (GTC). FRIDA will be the first GTC instrument to use the telescope's adaptive optics system.
Image from IAC
It is conceived as a flexible platform to integrate PELEA into a Multi-agent System using the MAGENTIX technology. PELEA is a planning-execution architecture designed both to simulation and real execution. The new model includes this functionality to perform total simulations, partial simulations or real executions.
The Roberry will be an autonomous rover, using the Hackberry A10 board, that can run missions onboard. Roberry has onboard planning and execution capabilities, which means it can generates and execute plans by itself. It is planner independent. It can receives both a goals list or a preloaded plan through WIFI.
Planning, Execution and LEarning Architecture (PELEA) is a domain-independent online planning and execution architecture which is conceived as a general-purpose architecture suitable for problems ranging from robotics to emergency management. It is also intended to provide a rapid prototyping life-cycle for building planning applications and support planning practitioners with a highly-configurable tool.
It is a tourist recommendation and planning application to assist users on the organization of a leisure and tourist agenda. The system offers the user a list of the city places that are likely of interest to the user. With a planning module schedules the list of recommended places according to their temporal characteristics as well as the user restrictions; that is the planning system determines how and when to perform the recommended activities.
The systems helps the user to get customized plans according to its tastes and preferences.