Тёмный

Agus Sudjianto on Machine Learning, AI, and Validation 

Dimitri Bianco
Подписаться 49 тыс.
Просмотров 1,4 тыс.
50% 1

I had another great conversation with Agus Sudjianto around machine learning models, LLMs, risk management, model validation, and building culture. Agus recently retired from Wells Fargo and is now enjoying a variety of side projects including working at H2O.ai, teaching model validation at UNC, and I believe a few other projects which may come to light within the next year.
Website:
www.FancyQuant...
Support the channel:
ko-fi.com/fanc...
Quant t-shirts, mugs, and hoodies:
www.teespring....
Connect with me:
/ dimitri-bianco
/ dimitribianco

Опубликовано:

 

15 сен 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 6   
@masteroogway8601
@masteroogway8601 10 дней назад
Its nice to hear javanese accent, proud of you pak agus
@fireinthehole2272
@fireinthehole2272 12 дней назад
Amazing interview, I loved the segment on responsible AI and how validation shows you when and how the models will fail. This is so awesome! Where can one learn the concepts, or is it a thing you learn on the job?
@DimitriBianco
@DimitriBianco 10 дней назад
@fireinthehole2272 it's typically leaned on the job. Really the questions you are answering are, if an assumption of the model is broken, what will happen.
@Clonehunterfighter88
@Clonehunterfighter88 13 дней назад
So is resampling for testing models useless? Given that you’re pulling from the same distribution? What are some solutions for model robustness?
@DimitriBianco
@DimitriBianco 13 дней назад
@Clonehunterfighter88 in finance you need to use a time split instead of random sampling. For example, training data could be 2018-2021, 2022 is validation, and 2023 is testing. Getting a model to work in all those scenarios is hard however it helps test the robustness of the model as it is used across time.
@Jamingooo
@Jamingooo 9 дней назад
Hey Dmitri, what do you think the future is looking like for quant trading in the next 10 years? Also, my mind is tech oriented, I feel like this is what I was born to do, but with recent layoffs and recession I don’t know if I can go to college to study it. I don’t want to be a lawyer or a doctor. Also, I had electrical engineer friend who’s from Los Angeles, he couldn’t get a job for a year, so he moved to North Carolina out of desperation because that’s where he got the job offer from. Another friend who’s just graduated with a mechanical engineering degree works at Starbucks. I feel like pursuing engineering is not an option either. On a cherry on top, people say quant is super competitive and there are more graduates than jobs available. So my question is, what jobs can I pursue that are in demand that perhaps are related in math, computer science, geometry etc. What can pay me at least $30-40 an hour, and not oversaturated?
Далее
Stan Uryasev's Quantitative Finance Journey
58:43
Просмотров 2,1 тыс.
Russian soldiers get chased by Ukraine drone
00:28
Просмотров 55 тыс.
Sal Khan: How AI Will Revolutionize Education
1:09:12
Просмотров 109 тыс.
2 years in Dubai - my honest thoughts
16:13
Просмотров 456 тыс.
ICML 2024 Tutorial: Physics of Language Models
1:53:43
Просмотров 18 тыс.
Ruby on Rails: The Documentary
44:16
Просмотров 253 тыс.
Optimal Protocols for Studying & Learning
1:41:39
Просмотров 810 тыс.
How AI was Stolen
3:00:14
Просмотров 894 тыс.
AI and Quantum Computing: Glimpsing the Near Future
1:25:33
What Is an AI Anyway? | Mustafa Suleyman | TED
22:02
Просмотров 1,5 млн