michaelbloem

Applied scientist at Amazon. Interested in #ORMS #MachineLearning #Python #Rstats #DataViz

The postings on this site are my own, not my employer's

[michaelbloem @michaelbloem]

"DeepStock: Reinforcement Learning with Policy Regularizations for Inventory Management"

#orms #optimization

https://t.co/nzWsk0MAgJ

[michaelbloem @michaelbloem]

OptiGuide: Large Language Models for Supply Chain Optimization

#ORMS

https://t.co/pNxaUN0ELn

[michaelbloem @michaelbloem]

Consensus Planning with Primal, Dual, and Proximal Agents

#orms

https://t.co/qippuLoEKF

[Andrej Karpathy @karpathy]

3e-4 is the best learning rate for Adam, hands down.

[michaelbloem @michaelbloem]

"Regularized Online Allocation Problems: Fairness and Beyond" by Balsiero, Lu, and Mirrokni

#ORMS

https://t.co/IEUyMjfszO

[michaelbloem @michaelbloem]

"Stockpyl is a Python package for inventory optimization and simulation."

#Python #ORMS

https://t.co/Q3vfsHvXQP

[michaelbloem @michaelbloem]

"Learning to learn learning-rate schedules"
from @AmazonScience

#MachineLearning #ORMS #deeplearning

https://t.co/nOuXS9azwC

[Google AI @GoogleAI]

RGD, a new variant of the stochastic gradient descent algorithm, re-weights data points during each optimization step based on their difficulty. Learn how this reweighting algorithm can boost performance on unseen data across various learning tasks → https://t.co/9gN1cUpCZC

[michaelbloem @michaelbloem]

"Oracle-Structured Bundle Distributed Optimization (OSBDO)"

#ORMS

https://t.co/Whk17QjLQw

[michaelbloem @michaelbloem]

This sounds great, especially for "nit" comments :-)

"Resolving code review comments with ML"

#ArtificialIntelligence #softwaredevelopment

https://t.co/mEsXinSsv5

[michaelbloem @michaelbloem]

"A Generalist Dynamics Model for Control"
from @DeepMind

#ORMS #ArtificialIntelligence

https://t.co/Dw7oc6Ifgl

[michaelbloem @michaelbloem]

Northwestern University Open Text Book on Process Optimization

#ORMS

https://t.co/KjsRBnv4ay

[michaelbloem @michaelbloem]

"On Dual-Based PI Controllers for Online Allocation Problems"

I love the connection between classic PI control and (online) optimization.

#ORMS

https://t.co/5Hrd1HTnWt

[michaelbloem @michaelbloem]

Deep Inventory Management

"This work provides a Deep Reinforcement Learning approach to solving a periodic review inventory control system with stochastic vendor lead times, lost sales, correlated demand, and price matching."

https://t.co/jHNeYMyg2C

#ReinforcementLearning

[michaelbloem @michaelbloem]

Google AI Blog: "Robust Online Allocation with Dual Mirror Descent"

https://t.co/i0Ps0MMFBm

via @googleai

#MachineLearning #ORMS

[michaelbloem @michaelbloem]

"A Deep Q-Network for the Beer Game: Deep Reinforcement Learning for Inventory Optimization"

paper: https://t.co/POC3R3RKPj

code: https://t.co/Al50HqiaOe

#MachineLearning #ORMS

[michaelbloem @michaelbloem]

"OptFormer: Towards Universal Hyperparameter Optimization with Transformers"

#MachineLearning

https://t.co/4HkjKtw07L

[michaelbloem @michaelbloem]

"Visual explanations of core machine learning concepts"

#MachineLearning

https://t.co/0X4k7P5Eo0

[michaelbloem @michaelbloem]

"Automatic termination for hyperparameter optimization"

#MachineLearning

https://t.co/Qi25DxGp6p

[michaelbloem @michaelbloem]

"MIP formulations and linearizations: quick reference"

A handy resource from @FICO Xpress

#ORMS #Optimization

https://t.co/aBAwdQDJlJ