I think this course is in the middle of a simple introduction and a practical course. You should not enroll if you expect to be able to be able to build a RL system. You should not enroll if you are expecting some simple intuitive introduction of RL. This is more difficult than an introduction but tells you nothing more than some introduction, so it is an introduction done in a difficult way. I think it is better to avoid it.

Only learned small pieces of concepts about quant trading, reinforcement learning parts are not connected well at all, it's all about advertising Google Cloud services.

Disapponting.

Last project week 3 does not have any connection to the topic.

Most of week 3 lessons are hand waving general recommendations, not real teaching or discussions

I feel deceived.

Really general level concepts and does not go deep into the code of reinforcement models. The labs are scarce and not helpful at all.

I do not recommend this course to my friends.

Exercises are not prepared to help learners to understand ML for Trading.

It was ... OK. The lectures by the NYIF guy were immediately relevant to me, worth taking the course for. They should just have removed the Google stuff entirely and just started with an assumption of a basic knowledge of ML - just focus on the financial applications. So, bottom line: the good content is good, but mixed with a bunch of generic, time-wasting junk... that at least can be skipped over.

This course seemed like movie trailer where there many jargons are introduced which are definitely worth but the information on the same is very limited which does not make students comfortable.

This course was more towards introducing the facility in Google Cloud than on the Title of the course.

Very unusual course. Some useful theory on RL but very little practical coded examples of RL for trading. Heavy on pushing Google cloud services.

The content was not bad, however it was really oriented towards promotion of GCP services.

Also, there was no tutorial how to really develop a strategy with reinforcement learning ( only few advices).

If you expect to find a working example of RL trading bot using some exchange API and executing orders and so on - you won't find any. The part on RL algos is good (assuming you have good fundamental preparation on RL). But the examples are NOT from trading - good ole CartOpole... The part on trading is just on general theory of market risk, and not on RL trading. Don't waste your time on this course

No practical, and useless for people who only wants more details about implementation of RL algo in trading rather than details about GCP.

It was easy to follow but not easy. I learned a lot and I now have the confidence to implement Reinforcement learning to my own FX trading strategies. Thank you so much.

I learned new perspectives of trading - great

I've given a rating of 3 to this course because even it gives you an intuitive understanding of Reinforcement learning, it won't make you build your own RL agent in a trading environment, which is bad because you should be able to apply the theory that is thaught to you. That would give you even a deeper and more practical understanding of the subject. The most interesting part of this course, even if it was treated lightly, is the "Investment and Trading Portfolio Optimization". Of course, you could deepen the subjects that are taught in this part, but I would have liked it to have been dealt with more thoroughly.

Really good stuff, thank you! The Deep Q networks were a bit over my head, I will need to keep studying. It was good theory, but I would have like to see these models trade in the markets to really understand how they act in live trading environments.

Very knowledgable theories from Jack Farmer and the AutoML lab was quite straight forward. However, it would've been good to have the week 3 Portfolio Risk Management code added included as an actual lab exercise instead of talking through it.

Great introduction to some very interesting concepts. Lots of hands on examples, and plenty to learn

Material is a bit of a mix - the content repurposed from other GCP courses doesn't really mesh that well. Last lab is a bit of a disappointment - there's only really one way to approach it given the time available, and it doesn't give us the time to experiment with other ideas. Would've been nice to have e.g. 24 hours for this lab, but that'd probably be considerably more expensive. That said, I got what I wanted out of the course overall, which was a background in DRL that I could apply to my trading

perhaps an applied trading notebook would have been nice...I understand that liability issues might have arisen, but there might have been a reasonable avenue with repeat disclaimers, etc

Good course introducing concepts in RL. Wish course provided more examples of using RL in stock prediction.

It has good practical stuff, BUT not any practical RL related to trading.

Most of the course is a generic lecture about RL and LSTM taken from other courses. The rest is mostly advertisement for GoogleCloud, which it is not useful since you could do all exercises on a local laptop. Only a fraction of the course talks about finance and it is so generic that cannot be applied to any real world case.

The learning curve is broken. It's like teach you 1+1=2 first, then you need to do calculus yourself, and lecturer say "see, it's easy" and move on to deep neural network now......

It's an exceptionally difficult task to predict financial time series, and even harder to design an automated trading methodology that can take into account those forecasts while monitoring the trading environment (trading costs, other traders, sentiment). This final course is an ambitious attempt to expose learners to the most advanced concepts in the field.

To be able to comprehend the Reinforcement Learning materials appears to require expertise in deep learning far beyond sequential models, and also appears to need the volume and integrity of data only available to high-frequency trading firms. Thumbs up to the specialization curators for providing a non-trivial introduction.

The module that rescued the course (and lifted my rating to 5 stars) was the AutoML demonstration. I was reassured to see that Gradient Boosted Trees were chosen as the appropriate methodology, as this is what I have casually observed as being the most effective methodology in use today for end-of-day data problems. Looks like an amazing product, if you have the money!

After the first two courses, this one grabs you into the reinforcement learning spectrum. This topic has been revealing to me and its applications to trading