Tagged: ML

IEEE CQR 2018 – AI CoC Session


IEEE CQR-ETR 2018: “Discuss and identify the RAS (Reliability, Availability and Serviceability) challenges, requirements and methodologies in the emerging technology areas like the Cloud Computing, Wireless/Mobility (with focus on 5G technologies), NFV (Network Functions Virtualization), SDN (Software Defined Networking), or similar large-scale distributed and virtualization systems.”

“Discuss the RAS requirements and technologies for mission-critical industries (e.g., airborne systems, railway communication systems, the banking and financial communication systems, etc.), with the goal to promote the interindustry
sharing of related ideas and experiences. Identify potential directions for resolving identified issues and propose possible solution.”


Session Title: A Programmatic Approach for an Artificial Intelligence Code of Conduct.

Today’s DX, Digital Transformation, programs are all the rage, but it takes a fair amount of double clicking and inquisitive questioning to separate facts from vaporware. DX typically involves a wide variety of game changing initiatives intersecting analytics, automation, programmability, software-defined systems, end-to-end integration, service-level composition and controls… all coming together to optimize for Quality as a differentiated and value-based Human Experience. Therefore, Customer Delight metrics (rather than outmoded customer satisfaction ones) are set to redefine the “Q” in CQR, Communications Quality & Reliability in 5G.

While the Telecoms industry rallies toward a zero touch paradigms, which some happen to position as a Human-“OFF”-the-Loop panacea, this session will expose the need for considering, and possibly pivoting, to the kind of Operational Excellence that can only be delivered by fast adapting HMS, Human-Machine-Systems instead. Note the rise of Dataviz (Data Visualization,) ML’s (Machine Learning’s) Collaborative Filtering, AI’s (Artificial Intelligence’s) RecSys (Recommender Systems) and a fresh take on Cybernetics… which are driving innovation in HILT and HOTL (Human-“IN”-The-Loop and Human-“ON”-the-Loop, Computing, as well as delivering effective mass-personalization with Affective Computing powered by Human Dynamics’ analytics.

Once upon a time… Telecoms’ pioneered HFE, Human Factors Engineering: a holistic systems engineering discipline addressing people (culture, workstyle, skills,) processes (procedures, methods, practices,) and technologies (crafts, tools, systems) so that we can best humanize technology and make a compelling difference across the value chain and at all levels. Unfortunately, HCD, Human-Centered-Design, fell out of favor over time while, paradoxically, took off in emerging technology sectors under other disciplines. Lost in oblivion, louder but siloed voices inflicted the sort of self-defeating Technical Myopia that props up complexity and negates differentiated Quality Experiences. Today’s telecoms industry is impacted by disintermediation and commoditization as a result and, equally telling, keeping extremely busy with importing practices from sectors frowned at in a no so distant past.

We are now embarked on a new journey. The sought after outcome of any Digital Service Provider, DSP, is to be instrumental to Digital Citizens’ Quality Experiences with new experimentation, monetization and growth models. This takes agility and dynamic system-wide (horizontal and vertical) behaviors, which prompts effortless operability at unprecedented speed, scale and scope. Our work permeates design, development, delivery and serviceability, as intertwined and continuous lifecycles instead of lock-step waterfalls.

In this context, AI, Artificial Intelligence, enables us, humans, to envision and implement capabilities beyond the reach of legacy systems’ last gasps. By the same token, practices that might have appeared to serve us well in the past, are now becoming dysfunctional and latency-prone barriers. A successful path forward takes augmented Human-Machine Intelligence. Human-Centered-Design’s outcome oriented model calls for a programmatic approach of an AI’s Code of Conduct, so that we can best interface and collaborate… instead of making good on Elon Musk’s well know fears around AI.