Programme (Tentative)


  • Sunday, 21 August

    Registration & Opening Event

      • 12:00 - 14:00
        • Registration

          Rotunda

      • 14:00 - 16:00
        • Introduction to ML using JAX

          2 amphi bois + A201 + A301 Kale-ab Tessera

      • 16:00 - 16:30
        • Coffee Break

          Rotunda

      • 16:30 - 18:00
        • Refresher: Maths for Machine Learning

          Main Planery Elizavita Semenova

      • 20:00 - 22:00
        • Gathering and Evening Party

          TBA

  • Monday, 22 August

    Indaba Day 1

      • 07:30 - 08:30
        • Morning Coffee

          Rotunda

      • 08:30 - 09:00
        • Keynote: Opening Remarks + Welcoming

          Main planery Zohra Slim+ Karim Beguir

      • 09:00 - 10:30
        • Keynote: On the future of NLP

          Main planery Sebastian Ruder

      • 10:30 - 11:00
        • Parallel Interactive Parallel Tracks

          TBA TBA

      • 11:00 - 12:30
        • Parallel Intro to Causality

          Amphi Rouge St John Grimbly

          If you’ve ever taken a statistics class, I have little doubt you’ve heard someone say “correlation does not imply causation”. This trope seems to imply that we can never state that A causes B by analysis of data alone. This talk will take a deeper look at if this is always the case by introducing the field of causal inference, with a special focus on (graphical) causal modelling. Attendees will leave having gained knowledge of how assumptions and data come together to enable drawing causal conclusions from data. More importantly, they will develop the basic intuition and fundamental concepts of causal modelling in the context of machine learning.

        • Parallel Few-shot Meta Learning

          Amphi Bois1 Jonathan Schwarz

          Despite groundbreaking successes of Machine Learning in the past decade, many practical problems remain outside the reach of modern approaches due the absence of large datasets. This poses a major challenge in numerous settings such as medical applications, robotics, low-resource languages and others. In this session, we will explore recent work on Meta- & Few-shot learning algorithms, which are explicitly trained for fast adaptation to small datasets. The objective of this session is to motivate the research area, introduce the fundamental building blocks of modern algorithms and provide you with a set of methods that have established themselves over the years. In addition, we will discuss open problems and challenges in the area and provide you with pointers for additional reading.

        • Parallel Array Algebra

          Amphi Bleu Taliesin Beynon

          Array algebra refers to the operations that we can perform on rectangular arrays (“tensors”) of numbers, which are the fundamental data structures used in deep learning, as well as more generally in data science and data-oriented programming. This talk, which pairs with the array algebra practical, will give a whirlwind tour of what arrays are, and help to organize the operations we typically perform on them. We’ll look at the natural sequence of n-arrays where scalars / numbers = 0-arrays, vectors / tuples = 1-arrays, matrices = 2-arrays, etc. We’ll review some concrete examples of different n-arrays, and what the axes of these arrays mean. We’ll triangulate among several different viewpoints of n-arrays, to gain a better intuition of the core array operations: generalized dot products, aggregation, reshaping, transposing, slicing, stacking, mapping, etc. We’ll also try to equip you with helpful ways of thinking about higher-order arrays and their axes, one of the more challenging aspects of coding with complex neural network architectures.

        • Parallel Monte Carlo 101

          Amphi Bois2 Steven James

          Many problems in machine learning require us to compute the sum or integral of a distribution. Examples of this include calculating the expected value of a random variable or computing the normalising factor in Bayes’ theorem. While there are certain cases where an analytic solution can be computed, most often an exact solution is intractable and so must be approximated.  In this talk, we cover Monte Carlo methods, where samples drawn from a distribution can be used to numerically approximate the quantity of interest. We describe rejection and importance sampling, which can be used even when the target distribution is not easy to sample from. However, these approaches do not work well when applied to huge models or datasets, as is commonly the case in machine learning. We therefore conclude with Markov Chain Monte Carlo – an algorithm that can be used to efficiently sample from high-dimensional distributions.

      • 12:30 - 14:00
        • Lunch Break

          Cafeteria

      • 14:00 - 16:00
        • Parallel Practicals: GNNs

          Amphi Bois1, A301 Matthew Morris

        • Parallel Practicals: Array Algebra

          Amphi Bois2, A201 Taliesin Beynon

      • 16:00 - 16:30
        • Parallel Interactive Parallel Tracks

          TBA

      • 16:30 - 18:00
        • Keynote: Plenary (Indaba Alumni Keynote)

          Main planery Aya Salama, Elizabeth Benson, Marcellin Atemkeng Teufack, Kaleab Abebe Tessera, Afonja Tejumade Mariam

      • 18:00 - 20:00
        • Network event (Discussion on AI in Africa)

          library and main hall

      • 20:00 - 22:00
        • Mentoring: Light talk: Effective Applications

          library and main hall

  • Tuesday, 23 August

    Indaba Day 2

      • 07:30 - 08:30
        • Morning Coffee

          Rotunda

      • 08:30 - 10:30
        • Parallel Practicals: Reinforcement Learning

          A301, Amphi Bois1 Juan Claude Formanek

        • Parallel Practicals: Deep Generative Models

          A201, Amphi Bois2 James Urquhart Allingham

      • 10:30 - 11:00
        • Coffee Break

          Rotunda

      • 11:00 - 12:30
        • Keynote: Visual Domain Adaptation in the Deep Learning Era

          Main planery Gabriela Csurka

      • 12:30 - 14:00
        • Lunch Break

          Cafeteria

      • 14:00 - 16:00
        • Parallel Bayesian Inference

          Amphi Rouge Javier Antorán Cabiscol

          Bayesian inference provides us with a consistent rule with which to update our beliefs when we observe new data. Compared to the more common loss minimisation approach to learning, Bayesian methods offer us calibrated uncertainty estimates, resistance to overfitting, and even approaches to select hyper-parameters without a validation set. However, seemingly more involved mathematics and larger computational cost make Bayesian methods niche within the machine learning community.

          In this talk, we will explore the foundations of Bayesian inference. We will start by discussing how Bayesian inference is inherent to human learning and reasoning. We will then codify this intuition by delving into how a Bayesian model is constructed. Finally, we will discuss practical algorithms for Bayesian machine learning and applications.

        • Parallel Multi-Agent Reinforcement Learning

          Amphi Bois1 Arnu Pretorious

          In this talk, we will introduce the basic concepts of multi-agent reinforcement learning (MARL) and motivate why MARL is a useful approach to building large-scale decision-making AI systems. In particular, we will focus on value-based methods, from simple independent learning strategies to more advanced value-decomposition methods. This will showcase one line of research attempting to solve key multi-agent challenges. Finally, we will briefly discuss other efforts at the forefront of research and provide resources for those interested in pursuing the topic further.
        • Parallel Introduction to Transformers

          Amphi Bleu Ruan van der Merwe

          The transformer architecture introduced in 2017 has significantly impacted the deep learning field. This tutorial will cover the transformer in great detail so that you can:
          • Explain it to your CEO
          • Train a fellow machine learning engineer
          • Maintain a conversation with a seasoned transformer researcher

          We will also demonstrate how to test if they suit your problem quickly and, if they do, how to structure your problem and data to get them working.

        • Parallel TIme Series Models

          Amphi Bois2 Samaneh Kouchaki

          Ever-increasing qualities and quantities of data are routinely collected concerning all aspects of our life.  Developments in wearable sensors, smart home technologies, and the Internet of Things provide the industry with opportunities to improve its services. However, analysing time-series data collected in different applications poses several challenges as the data can have substantial artefacts, might be incomplete, and contain high variations. This tutorial will introduce several machine learning techniques to tackle these issues and provide robust solutions to address such challenges. Various real-world examples especially on dealing with time series data in healthcare settings will also be provided.

      • 16:00 - 16:30
        • Coffee Break

          Rotunda

      • 16:30 - 18:30
        • Parallel Practicals: Bayesian Deep Learning

          Amphi Bois1, A301 Javier Antorán Cabiscol

        • Parallel Practicals: Transformers and Attention

          Amphi Bois2, A201 Ruan van der Merwe

      • 18:30 - 20:00
        • Dinner

          Cafeteria

      • 20:00 - 22:00
        • Parallel Communicating your research

          library/ main hall

        • Parallel Apple side event

          amphiater + main hall

  • Wednesday, 24 August

    Research in Africa Showcase

      • 07:30 - 08:30
        • Morning Coffee

          Rotunda

      • 08:30 - 09:30
        • Keynote: Contextualizing AI for social good on the African Continent

          Main planery Ernest Mwebaze

      • 09:30 - 10:00
        • Poster Research Spotlight Talks l

          Posters Area Multiple

      • 10:00 - 12:30
        • Poster and Demo sessions l

          posters Area

      • 10:30 - 11:00
        • Coffee Break

          posters Area/library

      • 12:30 - 14:00
        • Lunch Break

          Cafeteria

      • 14:00 - 15:00
        • Group Photo Session

          TBA

      • 15:00 - 15:30
        • Poster Research Spotlight Talks ll

          posters Area/library Multiple Speakers

      • 15:30 - 17:00
        • Poster and Demo sessions ll

          Posters Area

      • 16:00 - 16:30
        • Coffee break

          posters Area/library

      • 17:00 - 18:30
        • Parallel Lacuna Fund: Agricultural Datatsets and NLP Datatsets Zindi Turtle Hackathon

          TBA

        • Parallel "how I did it" 15 minute talk

          TBA

        • Parallel Mentorship session

          TBA

      • 18:30 - 20:00
        • Dinner

          Cafeteria

      • 20:00 - 22:00
        • Parallel Light talk: Navigating Acadamic/Industry Careers

          library/ main hall

        • Parallel instadeep side event

          TBA

  • Thursday, 25 August

    Workshops

      • 07:30 - 08:30
        • Indaba WIML event + Morning Coffee

          TBA

      • 08:30 - 10:30
        • Parallel NLP Workshop

          Amphi Bois1

        • Parallel All about Startups

          Amphi rouge Karim Beguir

        • Parallel AI Alignment

          amphi bleu Fazl Barez

      • 10:30 - 11:00
        • Coffee Break

          Rotunda

      • 11:00 - 12:30
        • Keynote: Plenary

          Main planery Sara Hoker

      • 12:30 - 14:00
        • Lunch Break

          Cafeteria

      • 14:00 - 16:00
        • Parallel AML Health Workshop

          Amphi Rouge Chris Fourie

        • Parallel NLP Workshop

          Amphi Bois1 Wilhemina Nekoto

        • Parallel Workshop on Weakly-supervised Computer Vision

          Aphi Bois2 Raoul de Charette

          Website: https://wscv-indaba.github.io/

        • Parallel From Specialists to Generalists: The World of Generalised RL Agents

          Amphi Bleu Tamlin Love

      • 16:00 - 16:30
        • Coffee Break

          Rotunda

      • 16:30 - 18:30
        • Parallel AML Health Workshop

          Amphi Rouge Chris Fourie

        • Parallel NLP Workshop

          Amphi Bois1 Wilhemina Nekoto

        • Parallel Workshop on Weakly-supervised Computer Vision

          Aphi Bois2 Raoul de Charette

          https://wscv-indaba.github.io/

        • Parallel From Specialists to Generalists: The World of Generalised RL Agents

          Amphi Bleu Tamlin Love

      • 18:30 - 20:00
        • Dinner

          Cafeteria

      • 20:00 - 20:45
        • Ideathon Pitches

          amphi rouge Amal Rannen

  • Friday, 26 August

    Workshops & Closing

      • 07:30 - 08:30
        • Morning Coffee

          Rotunda

      • 08:30 - 10:30
        • Parallel ML at the Edge

          Amphi Bleu Sara Hoker

        • Parallel African-Francophone in AI, challenges and opportunities

          Amphi Bois2 Salomon Kabongo

        • Parallel Trustworthy AI

          Amphi Bois1 Girmaw Abebe Tadesse

      • 10:30 - 11:00
        • Coffee Break

          Rotunda

      • 11:00 - 12:30
        • Keynote: Plenary

          Main Planery Elaine Nsousie

      • 12:30 - 14:00
        • Lunch break

          Cafeteria

      • 14:00 - 16:00
        • Parallel ML at the Edge

          Amphi Bleu Sara Hoker

        • Parallel African-Francophone in AI, challenges and opportunities

          Amphi Bois2 Salomon Kabongo

        • Parallel Trustworthy AI

          Amphi Bois1 Girmaw Abebe Tadesse

      • 16:00 - 16:30
        • Coffee Break

          Rotunda

      • 16:30 - 17:30
        • Panel Discussion

          Main Planery Rockerfeller / data.org

      • 17:30 - 18:30
        • Awards: Awards & Closing

          Main Planery TBA

      • 20:00 -
        • Closing Event Party

          TBA

  • Monday, 29 August

    AI Hack MEA (by registration only)

  • Tuesday, 30 August

    AI Hack MEA (by registration only)

  • Wednesday, 31 August

    AI Hack MEA (by registration only)

Please pick your parallel sessions by clicking on the one you like, and then hit the button below to download the pdf with your itinerary.