My Akwaaba to Akyiri story: A Personal Reflection of Deep Learning Indaba

Flora OladipupoFlora Oladipupo
10 min read

Journeying through the thoughts penned down by a witness of a wonderful event brings the need to see through the eyes of the writer and this entails understanding every choice of words used and this starts with the title of the article.

“Akwaaba” and “Akyirii” are expressions in Twi, a language spoken in Ghana, symbolizing “Welcome” and “Goodbye” respectively. “Akwaaba” conveys a warmth and welcoming spirit intrinsic to its meaning, while “Akyirii,” signifying farewell, was a word I was reluctant to embrace during this thrilling journey. "Yebetumi" is the theme for 2023 Indaba which means being capable and in that spirit of "we can"

Deep Learning Indaba (DLI) is the biggest Data Science and Machine Learning conference in Africa it is an event I have been looking forward to since I heard about it in 2022 though my application wasn't accepted then, that was more motivation for me to prepare ahead for 2023.

"Failure is not the end; it's just a better comeback called success waiting to happen."

There are four perspectives to this event for me which are:

  1. The Journey

  2. Learning

  3. Adventure

  4. Networking/Socials


THE JOURNEY

This event was a significant one for me because it wasn't just about attending DLI but also this would be my first opportunity to leave my birth country Nigeria to a different country. The journey from Nigeria to Ghana took about 20 hours going through two countries Benin Republic and Togo; marking the most extended road trip I've ever embarked on. While this sounds exhaustive it was an exciting one meeting people on the bus, seeing places, and marveling at some glaring differences from Nigeria like seeing how much more organized Bénin République, Togo, and Ghana actually looked which was far from what I expected based on how the places are perceived back home.

A prominent contrast between Nigeria and the three countries I got to see during the journey was the level of organization at these places. The moment we exited Lagos, Nigeria, the rugged roads transformed into smooth ones. In Bénin République, were a prevalent means of transport, utilized by both men and women alike. It was surprising to see traffic lights being obeyed by both vehicles and bikes and there was a distinct demarcation for bike riding on major roads which made it safe for bike riders. Togo, although small, is picturesque and an ideal vacation destination due to its numerous beaches. We reached Aflao, Ghana, at nightfall, so getting a glimpse of the country was challenging, and we still had approximately five hours of travel left to reach our final destination, Accra, the capital city of Ghana. However, I had an entire seven days to explore the country.

LEARNING

The Deep Learning Indaba conference spanned a total of seven days, which I utilized to the fullest extent possible. Contrary to the common perception of conferences being a series of dull lectures in a hall, Indaba distinctly stood out. The learning experiences were as exciting as the fun activities, with each moment offering something new to learn, whether it was through mentorship sessions, keynote talks, workshop sessions, or interactions with other attendees. The conference was rich with opportunities for gaining knowledge, making the learning process something I eagerly anticipated.

Keynote sessions

The Keynote lectures invariably provoked thoughtful reflection. The first keynote particularly resonated with me; it was titled "The Role of African Universities in developing intelligent systems for Africa and Beyond" by Ayokor Korsah. I found myself wishing all the lecturers from my school could be present to witness the remarkable initiatives being undertaken in the school to encourage students to embrace technology. I also greatly appreciated other keynote talks, including "Community Rooted AI Research" by Timnit Gebru, and "A Self-Serving Approach to Living a Good Life as a Researcher" by Rosanne Liu. These lectures revealed the extensive and varied work being conducted in the field, contributing to a rich and diversified learning experience.

Workshop sessions

The sessions that most resonated with my interests were hosted by the organization InstaDeep Ltd. The series commenced with a session on Machine Learning for Biology—a session I found particularly enlightening, given my background in Biology and undergraduate studies in Biotechnology. During this session, I was introduced to new concepts like ESM, a Large Language Model (LLM), and its utility in predicting protein sequences. Another session I found enlightening was focused on Reinforcement Learning, marking my initial in-depth exploration into this facet of Machine Learning. Overall, these sessions provided opportunities to learn, relearn, and unlearn various concepts and practices, which I intend to apply in the future.

Mentorship sessions

I made sure not to miss the mentorship sessions as they offered a unique opportunity to engage with experts in distinct fields of Artificial Intelligence. I had the chance to have one-on-one interactions and ask questions to Shekoofeh Azizi a research scientist at Google DeepMind, Tyna Eloundou a researcher at OpenAI, Daniela Massiceti a researcher at Microsoft, Iretiayo Akinola a researcher at Nvidia, Ibrahim Yusuf a researcher at InstaDeep. Conversations with these experts were revealing and illuminating. These interactions significantly contributed to shaping my trajectory towards my long-term career goals in Artificial Intelligence, providing clarity and insight into the field’s diverse possibilities.

Poster sessions

Yet another learning avenue for me was the poster presentation session. I had the opportunity to present my poster on Natural language Processing: The role it played in Nigeria's 2023 Presidential Election. It's an interesting project I and Abisola Agboola collaborated on. During my presentation, I received valuable insights from engaged audience members on how to expand the project and apply it to address the prevalent electoral malpractices in many African countries. This session, through exploring the works of others, also provided a glimpse into the diverse applications of AI, enriching my understanding of its multifaceted applications.

ADVENTURE

Friendship

This is my first time leaving Nigeria right so taking in a lot of views in this new environment is important right? And I did just that! However, before I delve into those details, I must highlight a significant aspect of my experience: meeting the individuals I now call friends, who were part of my adventurous journey in Ghana. They are Khaulat Abdulhakeem a Masters student at Stanford and content creator, Ali Zaidi a ML engineer at Microsoft and Stephen Paul a software engineer and GDSC lead. Our meeting may have been coincidental, but it felt like we had known each other for ages. They turned my stay in Ghana into a delightful and less isolated experience, providing companionship far away from home.

The Food

Before I immerse into the full recount of our adventure, I must express my thoughts on the food in Ghana. The culinary experience is something I’ve always been curious about, especially the ongoing ‘war’ surrounding the Ghanaian Jollof there’s always been a playful contention with Nigerians about whether Nigerian or Ghanaian Jollof rice is superior. Without donning the hat of a food critic, finding potato balls in a plate of Jollof rice instead of plantain was somewhat unusual to me, but it didn’t stop me from indulging.

I was eager to taste when we were served, yet, before even tasting, I noticed the absence of the smoky aroma inherent to Nigerian Jollof an aroma so pronounced, one would know of the dish’s presence upon entering a room. After the first spoonful of rice I knew its a big slap for Ghanaian jollof to be compared to Nigerian jollof and wondered who started the war in the first place because there was clearly a huge difference in the two taste. I initially thought the war existed because the dishes were similar, but after tasting, I’d readily choose a plate of concoction Jollof over Ghanaian Jollof any day. And this isn’t bias it’s a simple truth, a sentiment echoed by other attendees who had experienced both dishes. The meals I really enjoyed during my stay in Ghana was the potato, sauce, finger food and breakfast meals.

Sightseeing

Returning to the main narrative, we began our exploration by visiting the Accra mall, which, although quite nice, had relatively high prices. Then we went off to visit the Black star square and Independent Arch along the way I came across the most beautiful mosque i've ever seen. Upon reaching our destinations, we took really nice picture and content enough to make the moments imprinted in our memory

We also took time to visit a nearby beach, a tranquil retreat from the academic atmosphere of the conference, where we relished the sun's warmth and the ocean's breeze truly a moment of bliss. At the beach, we encountered some locals; surprisingly, one hailed from Mushin, Nigeria. Encountering someone from a familiar locale in this unfamiliar territory was a comforting experience. Skipping some conference sessions for these refreshing moments didn’t feel regrettable at all.

We concluded our day at a Jazz bar, savoring delicious food and mocktails to the soothing tunes of Frank Kissi’s jazz music. This was a unique bonding experience; engaging in diverse conversations, even though this was our initial meeting, forged a sense of closeness among us. It was then that I realized I yearn to recreate such experiences in various countries, with wonderful people I can consider friends. I am optimistic that such opportunities will materialize soon.

Networking/Socials

This mostly happened mostly at unexpected times during lunch, parties, workshop session and social night. Reflecting on it, it was during such times that I met my friends. I initially encountered Khaulat on the bus to Ghana, but our conversation didn’t spark until the welcome party, where we shared dances in the 360 photo booth. Ali and I crossed paths at lunch, discovering common ground over his employment at Microsoft and my involvement as a Microsoft Learn Student Ambassador. Stephen and I had already initiated conversations while still in Nigeria, despite our first face-to-face meeting occurring in Ghana.

Networking constituted a crucial component of the Indaba experience for me, providing the opportunity to personally meet individuals whose work I had admired from a distance. I had the privilege of engaging in a one-on-one discussion with Jeff Dean a chief scientist at Google Deep mind during the Google Networking Event. I was invited to lunch with Rosanne Liu the co-founder of ML Collective and research scientist at Google Brain.

The MILA social event proved to be enlightening for me, offering clarity and direction for my post-academic pursuits as a recent graduate stepping into the field of AI. I am keen to explore the various opportunities unveiled during the event.

Additionally, I found myself immersed in spontaneous conversations with people from various countries. A notable interaction was with a young Algerian couple, in their early 20s, who had wed just a week before attending DLI and were planning to spend their honeymoon in Ghana. We delved into discussions about the challenges of marrying young in Nigeria, given the existing national conditions that already pose substantial disadvantages, and the quest for financial stability before considering marriage. Understanding diverse perspectives on such universal life subjects was profoundly enlightening.

One social night that stood out to me was the cultural night party. This occasion allowed representatives from the 22 African countries attending Indaba to exhibit their traditional attire. It's worth mentioning that participants from Nigeria and South Africa were predominant at DLI. I took pride in representing my Yoruba tribe, donning traditional 'iro and buba' attire. The evening was a visual delight, showcasing a unity among attendees despite our diverse ethnic backgrounds. A wonderful cultural representation was from Bunmi Akinremi who did a poem rendition representing Yoruba culture as an Akewi. During this event, I networked with Jeroen Van Goey a senior BioAI engineer at InstaDeep.

During this event, I had the opportunity to network with Jeroen Van Goey, a Senior BioAI Engineer at InstaDeep, who expressed interest in learning about my research work. The congregation of diverse cultures and the connections formed added a rich, enlightening layer to the experience, making it memorable.

Each greeting of “Akwaaba” in this adventure was shadowed by an inevitable “Akyirii,” each welcome a precursor to a farewell, each introduction a stepping stone to a departure. This served as a constant reminder of the fleeting yet significant essence of such profound interactions, a dance between excitement and resignation, embodied in these two meaningful Twi expressions. In every encounter, there was a whisper of "Yebetumi," a reminder that “we can,” an echo of possibility and potential that intertwined with our shared experiences and learnings.

The “Akyiri” or the farewell, I believe, is the most heartrending part of any journey. Parting with individuals I had met at the event was particularly challenging, especially with my newfound friends with whom bonds had formed through our shared adventures. Our separations were laden with emotions, each departure echoing a silent wish for more shared moments, more learned insights. There was a unified wish for DLI to extend its enriching embrace a little longer, but like all beautiful experiences, a conclusion is inevitable. The hope remains alive for reconnections at the next DLI, but until then, the bonds and conversations continue, bridging distances and keeping the shared memories vibrant.

Akyiri Ghana

20
Subscribe to my newsletter

Read articles from Flora Oladipupo directly inside your inbox. Subscribe to the newsletter, and don't miss out.

Written by

Flora Oladipupo
Flora Oladipupo

I am a Data scientist who enjoys working with datasets, solving problems and exploring Artificial Intelligence