Artificial Intelligence Conference in London 2018
2018-10-09..2018-10-11
London, United Kingdom
who it's for
machine learning engineers, data engineers, engineering managers, platform engineers, research scientists, product managers, ethics and compliance leads
topics
Amazon SageMaker platform; deep learning at scale; convolutional neural networks; synthetic data with GANs; explainability and bias detection; production model deployment and monitoring; computer vision for retail and warehouses; checkout-less store systems (Amazon Go); low-latency voice NLU (Alexa); data versioning and provenance (data as code); automated article generation with Wikidata
| title | author | length |
|---|---|---|
| AI in production: The droids you’re looking for | Jonathan Ballon | 06m |
Uses of edge AI move from sci fi droids to deployed systems: Hikvision retrofits a warehouse with autonomous vision robots shipping $15M/day while cutting… | ||
| The state of automation technologies | Ben Lorica, Roger Chen | 09m |
Recent evidence shows adoption is real but uneven: an 11,000-respondent survey finds many firms in early stages and skills gaps, Safari learning has 2.1… | ||
| AI and machine learning at Amazon | Ian Massingham | 06m |
Amazon describes two production wins: Amazon Go checkout-less stores use computer vision and sensor fusion to let customers walk out and get charged by push… | ||
| Why we built a self-writing Wikipedia | Amy Heineike | 05m |
Primer builds Quicksilver, a system that drafts first Wikipedia pages for missing scientists by linking names across news, Wikipedia and Wikidata and using… | ||
| Trust and transparency of AI for the enterprise | Ruchir Puri | 12m |
Claims a delivered IBM Cloud capability that wraps any model (AWS Google Microsoft or on‑prem) to detect bias at runtime, flagging skewed behavior as decisions… | ||
| AI for a better world | Ashok Srivastava | 04m |
Economic disparity gets attacked with a concrete product: QuickBooks Capital uses rich datasets and advanced modeling to forecast small‑business cash flow… | ||
| Rethinking software engineering in the AI era | Yangqing Jia | 03m |
Argues that conventional rule‑based feature engineering gives way to learned models, replacing hand‑crafted descriptors with convolutional neural nets trained… | ||
| The missing piece | Cassie Kozyrkov | 04m |
Claims that failure comes from confusing research AI vs applied AI and that businesses routinely hire microwave builders instead of cooks. Points out plentiful… | ||
| Deep learning at scale: A field manual | Jason Knight | 03m |
Argues that data is code and must get the same practices as software: versioning via immutable stores or copy‑on‑write snapshots, metadata captured for model… | ||
| Fireside chat | Marc Warner, Louis Barson | 03m |
Discussion lists concrete barriers: 80% of data not amenable to AI limits projects even on NHS crown‑jewel sets, and diversity and bias in those datasets skews… | ||
| Notes from the frontier: Making AI work | Michael Chui | 03m |
McKinsey Global Institute surveys show deployment counts: 20% in 2017 moving to 50% now of respondents saying AI is embedded in at least one business unit, and… | ||
| Building artificial people: Endless possibilities and the dark side | Supasorn Suwajanakorn | 03m |
A recurrent neural network trained on 14 hours of Obama video synthesizes new lip‑sync that matches arbitrary audio but only the mouth is synthesized while the… | ||
| interviewed at the Artificial Intelligence Conference, London 2018 – Susan Etlinger | Susan Etlinger | 28m |
Argues that current ML practice creates real customer harms unless fixed: supervised learning promotes the status quo illustrated by hiring models that train… | ||
| Accelerating AI algorithms | Gary Brown | 13m |
Intel announces low-power vision accelerators and a software path to move models from x86 to VPUs and FPGAs, plus a USB Intel Neural Compute Stick for local… | ||
| interviewed at the Artificial Intelligence Conference, London 2018 – Pete Warden | Pete Warden | 29m |
Cheap STM32 "Blue Pill" boards are programmed bare‑metal (OpenOCD via Raspberry Pi) using code on stm32_bare_lib GitHub and run on 20 kilobytes RAM, 64K flash… | ||
| interviewed at the Artificial Intelligence Conference, London 2018 – Amy Heineike | Amy Heineike | 27m |
Company builds NLP that drafts reports as well as a junior analyst, combining a hybrid template‑extractive‑abstractive pipeline with pointer models that can… | ||
| Using Amazon SageMaker to develop and deploy ML models | Ian Massingham | 19m |
Amazon SageMaker provides a full ML pipeline: data prep in Jupyter, scalable training on p3 GPU instances and the Deep Learning AMI, and production inference… | ||
| nothing matches | ||