Loading
Topic
Computer Vision
Last discussed
10mo ago
Dec 13, 2025
2 tagged in 12 months
The Mathematical Foundations of Intelligence [Professor Yi Ma]
Machine Learning Street Talk·Dec 13, 2025
All time
4
episodes
On record
2
took a position
Where they standWhere they stand
Talking pointsNot enough disagreement in the transcripts to form camps. These are the positions taken by people with demonstrated expertise on this topic first, then by how many people heard them on it, then by VoiceRank. Positions are extracted from transcripts by a model and may misattribute who said what. Every quote links to the episode it came from.
IM
Data augmentation is the secret and most important ingredient in making self-supervised learning work in computer vision.
“data augmentation is the secret and indeed most important ingredient in making self-supervised learning work so well”
Machine Learning Street Talk · Jun 2021 · 1 episode · 25K views on this topicYM
Current 3D reconstruction techniques in computer vision do not constitute true understanding, as they lack the structured, view-invariant world models that biological brains utilize.
“Have you noticed that when we see something in 3D we get excited because we understand the 3D. We understand the content. We parse it already our brain. But the machine has no idea what the heck is in there.”
Machine Learning Street Talk · Dec 2025 · 1 episode · 37K views on this topicExtracted by a model; may misattribute who said what.
In the transcripts
Episodes discussing this topic
Over time
Share of episodesThis topic's episodes as a share of every episode we analysed, in each complete month. We analyse more episodes every month, so a raw count would mostly show that growth. Recent months fill in as transcripts are analysed.
Complete months · first on record Jun 2021