Tecton
timezone
+00:00 GMT
SIGN IN
  • Home
  • Events
  • Content
  • Help
Sign In
Sign in or Join the community to continue

apply(recsys) Conference 2022 | Monolith: Real-Time Recommendation System With Collisionless Embedding Table

Posted Dec 12, 2022 | Views 230
# apply(recsys) 2022
# Production Use Case
# Model training
Share
SPEAKER
Youlong Cheng
Youlong Cheng
Youlong Cheng
Engineering Leader @ ByteDance (developer of TikTok and Douyin)

Youlong is an engineering leader at Bytedance, the company which develops social networking apps like TikTok and Douyin. His team mainly focuses on ML infrastructure for recommendation modeling and foundation algorithm research. Before Bytedance, Youlong worked in the Google TensorFlow team as a Tech Leader Manager.

+ Read More

Youlong is an engineering leader at Bytedance, the company which develops social networking apps like TikTok and Douyin. His team mainly focuses on ML infrastructure for recommendation modeling and foundation algorithm research. Before Bytedance, Youlong worked in the Google TensorFlow team as a Tech Leader Manager.

+ Read More
SUMMARY

We’ll provide an introduction to Monolith, a system tailored for online training. Our design has been driven by observations of our application workloads and production environment that reflects a marked departure from other recommendations systems. Our contributions are manifold: first, we crafted a collisionless embedding table with optimizations such as expirable embeddings and frequency filtering to reduce its memory footprint; second, we provide an production-ready online training architecture with high fault-tolerance; finally, we proved that system reliability could be traded-off for real-time learning. Monolith has successfully landed in the BytePlus Recommend product.

+ Read More

Watch More

30
Posted Apr 12, 2022 | Views 670
# Feature Stores
# Production Use Case
# Systems and Architecture
44:35
Posted Dec 12, 2022 | Views 310
# apply(recsys) 2022
# Panel Discussion
# Production Use Case