# on 14-Jun-2022 (Tue)

Do you want BuboFlash to help you learning these things? Click here to log in or create user.

#### Annotation 7088889007372

 #causality #statistics πΌ[π | do(π‘), π = π§] vs πΌ[π | π = π§] πΌ[π | π = π§] simply refers to the expected value in the (pre-intervention) population where individuals take whatever treatment they would normally take ( π ). This distinction will become important when we get to counterfactuals...
status not read

#### Parent (intermediate) annotation

Open it
he intervention do(π‘) occurs. For example, πΌ[π | do(π‘), π = π§] refers to the expected outcome in the subpopulation where π = π§ after the whole subpopulation has taken treatment π‘ . In contrast, <span>πΌ[π | π = π§] simply refers to the expected value in the (pre-intervention) population where individuals take whatever treatment they would normally take ( π ). This distinction will become important when we get to counterfactuals in <span>

#### Original toplevel document (pdf)

cannot see any pdfs

#### Flashcard 7093198392588

Tags
#DAG #causal #edx #has-images
[unknown IMAGE 7093198916876]
[unknown IMAGE 7093193149708]

status measured difficulty not learned 37% [default] 0

#### pdf

cannot see any pdfs

#### Annotation 7095669361932

 #DAG #causal #inference In this book, we stress the need to take the causal question seriously enough to articulate it, and to delineate the separate roles of data and assumptions for causal inference. Once these foundations are in place, causal inferences become necessarily less casual, which helps prevent confusion. The book describes various data analysis approaches to estimate the causal eο¬ect of interest under a particular set of assumptions when data are collected on each individual in a population. A key message of the book is that causal inference cannot be reduced to a collection of recipes for data analysi
status not read

#### pdf

cannot see any pdfs

#### Annotation 7095670934796

 #DAG #causal #inference This is not a philosophy book. We remain agnostic about metaphysical concepts like causality and cause. Instead, we focus on the identification and estimation of causal eο¬ects in populations, that is, numerical quantities that measure changes in the distribution of an outcome under diο¬erent interventions.
status not read

#### pdf

cannot see any pdfs

#### Flashcard 7095673556236

Tags
#causality #has-images #statistics

Question
In defining the backdoor criterion (Definition 4.1) for the backdoor adjustment (Theorem 4.2), not only did we specify that the [...] set π blocks all backdoor paths, but we also specified that π does not contain any descendants of π
Answer
adjustment

status measured difficulty not learned 37% [default] 0

#### Parent (intermediate) annotation

Open it
In defining the backdoor criterion (Definition 4.1) for the backdoor adjustment (Theorem 4.2), not only did we specify that the adjustment set π blocks all backdoor paths, but we also specified that π does not contain any descendants of π

#### Original toplevel document (pdf)

cannot see any pdfs

#### Flashcard 7095675653388

Tags
#causality #statistics
Question
Definition 3.2 (What is a [...]?) A variable π is said to be a cause of a variable π if π can change in response to changes in π
Answer
cause

status measured difficulty not learned 37% [default] 0

#### Parent (intermediate) annotation

Open it
Definition 3.2 (What is a cause?) A variable π is said to be a cause of a variable π if π can change in response to changes in π

#### Original toplevel document (pdf)

cannot see any pdfs

#### Flashcard 7095678274828

Tags
#causality #statistics
Question

πΌ[π | do(π‘), π = π§] vs πΌ[π | π = π§]

[...] simply refers to the expected value in the (pre-intervention) population where individuals take whatever treatment they would normally take ( π ). This distinction will become important when we get to counterfactuals...

Answer
πΌ[π | π = π§]

status measured difficulty not learned 37% [default] 0

Open it

#### Original toplevel document (pdf)

cannot see any pdfs

#### Annotation 7095680109836

 #causality #has-images #statistics Consider the following general kind of path, where β Β· Β· Β· β denotes a directed path: π β Β· Β· Β· β π β Β· Β· Β· β π . Conditioning on π , or any descendant of π in a path like this, will induce collider bias
status not read

Open it

#### Original toplevel document (pdf)

cannot see any pdfs

#### Flashcard 7095681944844

Tags
#causality #statistics
Question

Assumption 3.2 (Minimality Assumption)

1. Given its parents in the DAG, a node π is independent of all its non-descendants (Assumption 3.1).

2. Adjacent nodes in the DAG are [...].

Answer
dependent

status measured difficulty not learned 37% [default] 0

#### Parent (intermediate) annotation

Open it
Assumption 3.2 (Minimality Assumption) 1. Given its parents in the DAG, a node π is independent of all its non-descendants (Assumption 3.1). 2. Adjacent nodes in the DAG are dependent.

#### Original toplevel document (pdf)

cannot see any pdfs

#### Flashcard 7095683779852

Tags
#causality #has-images #statistics

Question
If we condition on a descendant of π that isnβt a [...], it could unblock a path from π to π that was blocked by a collider. For example, this is the case with conditioning on π in Figure 4.13. This induces non-causal association between π and π , which biases the estimate of the causal effect
Answer
mediator

status measured difficulty not learned 37% [default] 0

#### Parent (intermediate) annotation

Open it
If we condition on a descendant of π that isnβt a mediator, it could unblock a path from π to π that was blocked by a collider. For example, this is the case with conditioning on π in Figure 4.13. This induces non-causal association between π a

#### Original toplevel document (pdf)

cannot see any pdfs

#### Flashcard 7095685614860

Tags
#Data #GAN #reading #synthetic
Question

[...] GAN - Summary

There are many ways of dealing with this problem using different levels of granularity. We presented one that can be easily extended to handle temporal sequences. Its main advantages are modularity and scalability as it can be extended to multiple granularity levels.

The main drawback of this approach is that it may become difficult to implement if the database contains too many one to many relations and nested hierarchical levels.

Answer
hierarchical

status measured difficulty not learned 37% [default] 0

#### Parent (intermediate) annotation

Open it
hierarchical GAN - Summary There are many ways of dealing with this problem using different levels of granularity. We presented one that can be easily extended to handle temporal sequences. Its main

#### Original toplevel document (pdf)

cannot see any pdfs

#### Annotation 7095690857740

 [unknown IMAGE 7095689547020] #ML_in_Action #has-images #learning #machine #software-engineering
status not read

#### pdf

cannot see any pdfs

#### Annotation 7095692692748

 #ML_in_Action #learning #machine #software-engineering Machine Learning Engineering in Action
status not read

#### pdf

cannot see any pdfs

#### Annotation 7095694265612

 #ML_in_Action #learning #machine #software-engineering I had some successes but many failures, and generally left a trail of unmaintainable code in my wake as I moved from job to job. Itβs not something that Iβm particularly proud of. Iβve been contacted by former colleagues, years after leaving a position, to have them tell me that my code is still running every day. When Iβve asked each one of them why, Iβve gotten the same demoralizing answer that has made me regret my implementations: βNo one can figure it out to make changes to it, and itβs too important to turn off.β Iβve been a bad data scientist. Iβve been an even worse ML engineer. It took me years to learn why that is. That stubbornness and resistance to solving problems in the simplest way created a lot of headaches for others, both in the sheer number of cancelled projects while I was at companies and in the unmaintainable technical debt that I left in my wake. It wasnβt until my most recent job, working as a resident solutions architect at Data- bricks (essentially a vendor field consultant), that I started to learn where I had gone wrong and to change how I approached solving problems.
status not read

#### pdf

cannot see any pdfs

#### Annotation 7095695838476

 #ML_in_Action #learning #machine #software-engineering Part 1 (chapters 1β8) is focused primarily on the management of ML projects from the perspective of a team lead, manager, or project lead. It lays out a blue- print for scoping, experimentation, prototyping, and inclusive feedback to help you avoid falling into solution-building traps
status not read

#### pdf

cannot see any pdfs

#### Annotation 7095697411340

 #ML_in_Action #learning #machine #software-engineering Part 2 (chapters 9β13) covers the development process of ML projects. With examples (both good and bad) of ML solution development, this section carries you through proven methods of building, tuning, logging, and evaluating an ML solution to ensure that youβre building the simplest and most maintainable code possible
status not read

#### pdf

cannot see any pdfs

#### Annotation 7095698984204

 #ML_in_Action #learning #machine #software-engineering Part 3 (chapters 14β16) focuses on βthe afterβ: specifically, considerations related to streamlining production release, retraining, monitoring, and attribution for a project. With examples focused on A/B testing, feature stores, and a passive retraining system, youβll be shown how to implement systems and architectures that can ensure that youβre building the minimally complex solution to solve a business problem with ML
status not read

#### pdf

cannot see any pdfs

#### Annotation 7095700557068

 #ML_in_Action #learning #machine #software-engineering The vast majority of the time, a project that fails to make its way to production for sustained utility has issues that are rooted in the very early phases. Before even a single line of code is written, before a serving architecture is selected and built out, and long before a decision on scalable training is made, a project is doomed to either cancellation or unused obscurity if planning, scoping, and experimentation is not done properly.
status not read

#### pdf

cannot see any pdfs