Deep Learning vs Machine Learning – Overview & Differences – Morioh

Machine learning and deep learning are two subsets of artificial intelligence which have garnered a lot of attention over the past two years. If you’re here looking to understand both the terms in the simplest way possible, there’s no better place to be..
— Read on morioh.com/p/78e1357f65b0

Agile Manifesto – Personal Reflection

Individuals and Interactions over Process and Tools

Stereotypical software developers are introverts, heads down, coding. Articulating where they are in the development lifecycle sometimes heavily relies upon tools for measuring progress such as JIRA, Product Backlog status of User Stories, e.g. “In Progress” with an Effort estimation of 3.

“Blocked” User Stories may require the implementation team to “break out of their shell” and work with their teammates to “unblock” Product Backlog items. It breaks people out of their comfort zone. We need to discuss options and opportunities for removing blockers. “All for One, and One for all”

Working Product over Comprehensive Documentation

Over a decade or so ago, the measure of my merit was the complete test coverage of requirements for software implementation. Back then I was a QA lead, and my focus was to make sure all use cases for the software under development had complete test coverage.

Requirements changes from our business through our business analysts must be vetted with the QA team so use cases/test cases must be updated to ensure coverage. Sometimes a dependency of one requirement had a ripple effect throughout the software, so lots of documentation updates were required. Milestone dates were in many cases fixed, so teams were squeezed to do more with less time.

Flash forward to today, and leveraging Agile principles, I breathe a slight sigh of relief. Iterating product delivery via sprints every 2 weeks is supremely better than attempting to traverse updates to Business Requirements Documents (BRD), and technical specs. User Stories in a Backlog are much more succinct, and in some cases, a bit more abstract leaving functionality open to some level of ambiguity and interpretation.

Sprint Close scrum ceremonies every two weeks with our Product Owner, the central mouthpiece for the definition of the software product helps define the path forward. Did we get it right? Where do we need to make changes? There is no substitute for an evolving product and accompanying dialog with our Product Owner.

Customer Collaboration over Contract Negotiation

Both sides of the aisle seem to agree, building a solution with iterative input from the customer enables the product vision to be realized far better than without frequent touchpoints.

Statements of Work (SoW) to engage 3rd party solutions integrators (SI) may be abstract in some way. Holding vendors accountable for loosely formed requirements is tenuous at best. Quibbling about he said, she said is a waste of time.

Fail fast, engage regularly and often with our [Business] Product Owner enables us to collaborate on a working solution. The focus is on the evolving product vision and not the paper trail.

Responding to Change over Following a Plan

A “last-minute” change request? It could push back our timelines and accompanying milestones. Dates can’t change, and teams need to absorb the changes, i.e. nights and weekends. Responding to incremental changes at a regular cadence is a sustainable life cycle.

Agile’s Watergate

A relic of the Waterfall model is the construct of a “gate” process. In order for a project to achieve a milestone, the project/solution would need to achieve certain criteria that would allow it to go to the next phase of the project. For example, going from solidifying requirements in a Business Requirements Doc (BRD) to the software implementation phase.

In Agile, we leverage the Product Owner (PO) and the Product Backlog to determine what gets done and when. A Product Backlog item (PBI) may cover the full lifecycle of a Feature, from requirements to implementation. The Product Owner dictates acceptance of the PBI based on the status/transparency of the Backlog, such as the criticality of the Bugs linked to the PBI. Product quality and implemented functionality are transparent to the PO, who will determine the next steps such as release the software, and/or go through another iteration/sprint. Iterations are a defined cadence agreed to by the implementation team and the Product owner, typically, 2-week sprints.

Agile, Hybrid Environments: Opportunities for Synergy

Epics, Features, Product Backlog Items, and Tasks are object types in a Backlog that enable the PO and the team to link objects and plan over multiple sprints. Epics or Themes of Sprints are “high level”, potentially strategic initiatives. Features roll up into Epics as a part of several sprints. Either Epics or Features may be high enough level to link to Psydo Project Milestones for a product roadmap of deliverables, and solicitation outside the team.

Aggregation of Product Backlog Items, Effort Estimations, roll up into Features, and then up into Epics, which roughly equate to milestone timelines.

The “Definition of Done” (DoD) for a Product Backlog Item may require 0 outstanding Bugs with the severity of “Critical” linked to this PBI. The DoD criteria could be analogous to a traditional Quality Assurance gate.

Tasks that are production rollout activities, without a project plan, should be planned for in future sprints, akin to estimating when items may be completed in the proper sequence. Some of the Tasks may be placed conservatively in “early” sprints and may require items to be “pushed forward” after each of the iterations.

When Stakeholders Collide

Requirements Expedition

Maybe you’ll meet them during the Project Kickoff. Maybe you’ll first hear from them during a biweekly Steering Committee. Or maybe you will first hear from them three months into the project at a quarterly meeting with the CIO and the rest of his portfolio. Maybe you will never hear from them directly.

The politics of requirements gathering and prioritization is a daunting process. I’m not going to drudge up all the stories and categorize them here because it’s a painful process.

Why are some of your milestones in your project plan:

• the milestone exists within someone’s year end evaluation

• the requirements of a milestone are so bipolar, they are bound to fail. Need a project to bucket the requirements to say “we tried”, and we can pin it to a project.

• backing into established project timelines based on expectations set at the highest levels, e.g. regulatory compliance

Legal and Compliance Stakeholders

Global representation of legal and compliance requirements are a dichotomy of legal precedence between jurisdictions.

Agile Product Owner verse Waterfall Stakeholder Committee(s)

Many a project managed using waterfall kept me balancing the needs and wants of Stakeholders from all walks of life, some exuberantly voicing their opinions regardless of their position of power, or lack therein. The Agile Product Owner (PO) is a relief of burden, a single mouthpiece of the business, which dictates backlog priority.

Does Agile make the requirements gathering and prioritization pain go away? Possibly. There are various implementations of Agile, hybrid situations, and there are lots of tools out there to help manage the Product Backlog (requirements). Another exercise, developing User Journeys, working with your Personas / actors to derive their story, that is telling and lots of fun.

Help Wanted: Civil War Reenactment Soldiers to Improve AI Models

I just read an article on Digital PC Magazine, “Human Help Wanted: Why AI Is Terrible at Content Moderation” which started to get my neurons firing.

Problem Statement

Every day, Facebook’s artificial intelligence algorithms tackle the enormous task of finding and removing millions of posts containing spam, hate speech, nudity, violence, and terrorist propaganda. And though the company has access to some of the world’s most coveted talent and technology, it’s struggling to find and remove toxic content fast enough.

Ben Dickson
July 10, 2019 1:36PM EST

I’ve worked at several software companies which leveraged Artifical Intelligence, Machine Learning to recognize patterns, correlations. The larger the data sets, in general, the higher the accuracy of the predictions. The outliers in the data, the noise, “falls out” of the data set. Without quality, large training data, Artificial Intelligence makes more mistakes.

In terms of speech recognition, image classification, and natural language processing (NLP), in general, programs like chatbots, digital assistants, are becoming more accurate because of their sample size, training data sets are large, and there is no shortage of these data types. For example, there are many ways I can ask my digital assistant for something, like “Get the movie times”. Training a digital assistant, at a high level, would be to catalog how many ways can I ask for “something”, achieve my goal. I can go and create that list. I could write a few dozen questions, but still, my sample data set would be too small. Amazon has a crowdsourcing platform, Amazon Mechanical Turk, which I can request they build me the data sets, thousands of questions, and correlated goals.

MTurk enables companies to harness the collective intelligence, skills, and insights from a global workforce to streamline business processes, augment data collection and analysis, and accelerate machine learning development.

Amazon Mechanical Turk: Access a global, on-demand, 24×7 workforce

Video “Scene” Recognition – Annotated Data Sets for a Wide Variety of Scene Themes

In silent films, the plot was conveyed by the use of title cards, written indications of the plot and key dialogue lines. Unfortunately, silent films are not making a comeback. In order to achieve a high rate of successful identification of activities within a given video clip, video libraries of metadata need to be created, that capture:

  • Media / Video Asset, Unique Identifier
  • Scene Clip IN and OUT timecodes
  • Scene Theme(s), similar to Natural language processing (NLP), Goals = Utterances / Sentences
    • E.g. Man drinking water; Woman playing Tennis
  • Image recognition, in the context of machine vision, is the ability of software to identify objects, places, people, writing and actions in images. Image recognition is used to perform a large number of machine-based visual tasks, such as labeling the content of images with meta-tags

Not Enough Data

Here is an example of how Social Media, such as Facebook, attempts to deal with video deemed inappropriate for their platform:

In March, a shooter in New Zealand live-streamed the brutal killing of 51 people in two mosques on Facebook. But the social-media giant’s algorithms failed to detect the gruesome video. It took Facebook an hour to take the video down, and even then, the company was hard-pressed to deal with users who reposted the video.

Ben Dickson
July 10, 2019 1:36PM EST

…in many cases, such as violent content, there aren’t enough examples to train a reliable AI model. “Thankfully, we don’t have a lot of examples of real people shooting other people,” Yann LeCun, Facebook’s chief artificial-intelligence scientist, told Bloomberg.

Ben Dickson
July 10, 2019 1:36PM EST

Opportunities for Actors and Curators of Video Content: Dramatizations

All those thousands of people who perform, creating videos of content that range the gamut from playing video games to “unboxing” collectible items. The actors who perform dramatizations could add tags to their videos indicating as per above, documenting themes for a given skit. If actors post their videos on YouTube or proprietary crowdsourcing platforms, they would be entitled to some revenue for the use of their licensed video.

Disclosure Regarding Flag Controversy

I now realize there are politics around Nike “tipping their hat” toward the Betsy Ross flag. However, when I referenced the flag in this blog post, I was thinking of the American Revolution, and the 13 colonies flag. I didn’t think the title would resonate with readers, “Help Wanted: Amerian Revolutionary war Reenactment Soldiers to Improve AI Models.”, so I took some creative liberty.

Smart Solutions

Skip to toolbar