My ML objective

I am an ML researcher that is passionate, really passionate, about machine learning. I want to develop models capable of extracting the most information from all available data sources. At the same time, I want these models to generalize. In this development process I want to learn the most and help humanity using these autonomous models.

Research interests

I think that to develop such general models it is necesary to learn various ML subdisciplines: generative models, ML architectures, RL, transfer learning, causal structure learning, etc. From a theoretical perspective, I have had the oportunity to research VAEs and GANs for representation learning and clustering. From an applied perspective, I have applied transformers, semi-supervised learning, and anomaly detection methods to times series (astronomical lightcurves).

I am interested in any methodology or models that can help me advance in this long research objective. More specifically, I am interested in unsupervised learning, semi-supervised learning, multi-modal approaches, efficient/model-based RL and causal discovery.

Biography

I graduated as a Master and an Engineer in Electrical Engineering from the University of Chile in 2021. From the same university, I also obtained the bachelors of Electrical, Mechanical and Computer Engineering. In 2019, I did a 6 month research internship at Harvard IACS with Pavlos Protopapas and we published MPCC in ECCV2020. I have also published a variety of papers mainly focusing in unsupervised and semi-supervised learning. My main contributions are theoretical ML research in VAEs and GANs, and applied ML in astroinformatics.

My interest in ML began in 2015, doing Andrew NG and Daphne Koller courses. In 2016 I joined Laboratory of Computational Intelligence (LCI lab), under the supervision of Pablo Estévez. In LCI lab and in many chilean institutes the main focus of research is astroinformatics, thanks to the clean chilean skies. In 2024, the Rubin Observatory built in the north of Chile is going to start streaming data and modern ML methods are necesary to process the huge amount of data. Since 2016, my main areas of research have been unsupervised and semi-supervised learning as they were promising alternatives to take advantage of the huge amount of unlabeled astronomy data.

Increasing efforts of chilean and foreign universities/institutes have lead to investment to automatically analyze the data generated by modern synoptic telescopes such as ZTF, ATLAS and the future Rubin Observatory. ALeRCE lead by Francisco Forster is the result of one of these collaborations and the only one in the southern hemisphere of the earth. Currently, I am working as a part-time ML engineer in ALeRCE creating training sets, lifting services and putting in production ML models. In addition, I also work part-time as research scientist in collaboration with Pablo Huijse.

Posts

The posts are mainly about my work that I have not published yet but in the future I plan to make post about other subjects that I like.