Domino Data Lab

AUGUST 6, 2021

With a framework and Enterprise MLOps, organizations can manage data science at scale and realize the benefits of Model Risk Management that are received by a wide range of industry verticals. It was first defined by the US Federal Reserve and Office of the Comptroller of the Currency ( SR 11-7 ) in April 2011. What Is Model Risk?

Risk Management 64

Risk Management Risk Modeling Enterprise 64

The Unofficial Google Data Science Blog

OCTOBER 12, 2017

We data scientists now have access to tools that allow us to run a large numbers of experiments, and then to slice experimental populations by any combination of dimensions collected. Make experimentation cheap and understand the cost of bad decisions. This leads to the proliferation of post hoc hypotheses.

Experimentation 61

Experimentation Testing Statistics Metrics 61

Occam's Razor

JUNE 4, 2018

Present a yummy spreadsheet that quantifies the cost of inaction , how much money you’ll lose by not delivering a 25% improvement every week. Yet, this incredible benefit was not a part of YouTube TV’s merchandizing strategy from day one. Checkout the Kimbao Sauvignon Blanc you can see sales and would buy it again rates since 2011.

Strategy 124

Strategy Cost-Benefit Testing Sales 124

Occam's Razor

NOVEMBER 2, 2015

The benefits are numerous. In my Oct 2011 post, Best Social Media Metrics , I'd created four metrics to quantify this value. Economic Value (EcV) is the value of short and long-term revenue and cost savings. The serious point is that when we choose to invest in Social Media, it comes at a cost. The opportunity cost.

B2B 167

B2B Metrics Testing Recreation/Entertainment 167

Domino Data Lab

AUGUST 22, 2019

During their data preprocessing steps, Maas and his coworkers decided to leave in stop words because they are “indicative of sentiment.” [Note: In more technical machine learning terms, the cost function of the skip-gram architecture is to maximize the log probability of any possible context word from a corpus given the current target word.]

Deep Learning 111

Deep Learning Modeling Metrics Testing 111

O'Reilly on Data

MAY 11, 2020

Television only lacked the immediate feedback that comes with clicks, tracking cookies, tracking pixels, online experimentation, machine learning, and “agile” product cycles. Regardless of how we think of ourselves, humans aren’t terribly good at trading off short-term stimulus against long-term benefits. And the cycle goes on.

Modeling 188

Modeling Advertising Recreation/Entertainment Metrics 188