Yannic Kilcher Videos (Audio Only)

#adversarialexamples #dimpledmanifold #security Adversarial Examples have long been a fascinating topic for many Machine Learning researchers. How can a tiny perturbation cause the neural network to change its output by so much? While many explanations have been proposed over the years, they all appear to fall short. This paper attempts to comprehensively explain the existence of adversarial examp...

#mlnews #gta #weather In this week's ML News, we look at the latest developments in the Machine Learning and AI world with updates from research, industry, and society at large. OUTLINE: 0:00 - Intro 0:20 - Hugging Face launches free course 1:30 - Sentdex releases GAN Theft Auto 2:25 - Facebook uses AI to help moderators 4:10 - Weather with Antonio 5:10 - Autonomous ship aborts mission 7:25 - PyTo...

#xcit #transformer #attentionmechanism After dominating Natural Language Processing, Transformers have taken over Computer Vision recently with the advent of Vision Transformers. However, the attention mechanism's quadratic complexity in the number of tokens means that Transformers do not scale well to high-resolution images. XCiT is a new Transformer architecture, containing XCA, a transposed ver...

#reiforcementlearning #gan #imitationlearning Learning from demonstrations is a fascinating topic, but what if the demonstrations are not exactly the behaviors we want to learn? Can we adhere to a dataset of demonstrations and still achieve a specified goal? This paper uses GANs to combine goal-achieving reinforcement learning with imitation learning and learns to perform well at a given task whil...

OUTLINE: 0:00 - Intro 0:30 - Google RL creates next-gen TPUs 2:15 - Facebook launches NetHack challenge 3:50 - OpenAI mitigates bias by fine-tuning 9:05 - Google AI releases browseable reconstruction of human cortex 9:50 - GPT-J 6B Transformer in JAX 12:00 - Tensorflow launches Forum 13:50 - Text style transfer from a single word 15:45 - ALiEn artificial life simulator My Video on Chip Placement:...

#implicitfunction #jax #autodiff Many problems in Machine Learning involve loops of inner and outer optimization. Finding update steps for the outer loop is usually difficult, because of the.need to differentiate through the inner loop's procedure over multiple steps. Such loop unrolling is very limited and constrained to very few steps. Other papers have found solutions around unrolling in very s...

#mlnews #wudao #academicfraud OUTLINE: 0:00 - Intro 0:25 - EU seeks to regulate AI 2:45 - AI COVID detection systems are all flawed 5:05 - Chinese lab trains model 10x GPT-3 size 6:55 - Google error identifies "ugliest" language 9:45 - McDonald's learns about AI buzzwords 11:25 - AI predicts cryptocurrency prices 12:00 - Unreal Engine hack for CLIP 12:35 - Please commit more academic fraud Referen...

#decisiontransformer #reinforcementlearning #transformer Proper credit assignment over long timespans is a fundamental problem in reinforcement learning. Even methods designed to combat this problem, such as TD-learning, quickly reach their limits when rewards are sparse or noisy. This paper reframes offline reinforcement learning as a pure sequence modeling problem, with the actions being sampled...

#mlnews #anthropic #eliza Anthropic raises $124M for steerable AI, peer review is threatened by collusion rings, and the original ELIZA source code was discovered. OUTLINE: 0:00 - Intro 0:40 - Anthropic raises $124M 3:25 - 65% of execs can't explain AI predictions 4:25 - DeepMind releases AndroidEnv 6:10 - Collusion rings in ML Conferences 7:30 - ELIZA's original source code discovered 10:45 - Ope...

#reinforcementlearning #deepmind #agi What's the most promising path to creating Artificial General Intelligence (AGI)? This paper makes the bold claim that a learning agent maximizing its reward in a sufficiently complex environment will necessarily develop intelligence as a by-product, and that Reward Maximization is the best way to move the creation of AGI forward. The paper is a mix of philoso...

#expirespan #nlp #facebookai Facebook AI (FAIR) researchers present Expire-Span, a variant of Transformer XL that dynamically assigns expiration dates to previously encountered signals. Because of this, Expire-Span can handle sequences of many thousand tokens, while keeping the memory and compute requirements at a manageable level. It severely matches or outperforms baseline systems, while consumi...

#fnet #attention #fourier Do we even need Attention? FNets completely drop the Attention mechanism in favor of a simple Fourier transform. They perform almost as well as Transformers, while drastically reducing parameter count, as well as compute and memory requirements. This highlights that a good token mixing heuristic could be as valuable as a learned attention matrix. OUTLINE: 0:00 - Intro &am...

#artificialintelligence #musicvideo #clip I used OpenAI's CLIP model and BigGAN to create a music video that goes along with the lyrics of a song that I wrote. The song lyrics are made from ImageNet class labels, and the song itself is performed by me on a looper. OUTLINE: 0:00 - Intro 1:00 - AI-generated music video for "be my weasel" 3:50 - How it was made 7:30 - My looping gear 9:35 - AI-genera...

#ddpm #diffusionmodels #openai GANs have dominated the image generation space for the majority of the last decade. This paper shows for the first time, how a non-GAN model, a DDPM, can be improved to overtake GANs at standard evaluation metrics for image generation. The produced samples look amazing and other than GANs, the new model has a formal probabilistic foundation. Is there a future for GAN...

#involution​ #computervision​ #attention​ Convolutional Neural Networks (CNNs) have dominated computer vision for almost a decade by applying two fundamental principles: Spatial agnosticism and channel-specific computations. Involution aims to invert these principles and presents a spatial-specific computation, which is also channel-agnostic. The resulting Involution Operator and RedNet architectu...

#mixer​ #google​ #imagenet​ Convolutional Neural Networks have dominated computer vision for nearly 10 years, and that might finally come to an end. First, Vision Transformers (ViT) have shown remarkable performance, and now even simple MLP-based models reach competitive accuracy, as long as sufficient data is used for pre-training. This paper presents MLP-Mixer, using MLPs in a particular weight-...

#genderbias​ #algorithmicfairness​ #debiasing​ A brief look into gender stereotypes in Google Translate. The origin is a Tweet containing a Hungarian text. Hungarian is a gender-neutral language, so translating gender pronouns is ambiguous. Turns out that Google Translate assigns very stereotypical pronouns. In this video, we'll have a look at the origins and possible solutions to this problem. OU...

#perceiver​ #deepmind​ #transformer​ Inspired by the fact that biological creatures attend to multiple modalities at the same time, DeepMind releases its new Perceiver model. Based on the Transformer architecture, the Perceiver makes no assumptions on the modality of the input data and also solves the long-standing quadratic bottleneck problem. This is achieved by having a latent low-dimensional T...

#universalcomputation​ #pretrainedtransformers​ #finetuning​ Large-scale pre-training and subsequent fine-tuning is a common recipe for success with transformer models in machine learning. However, most such transfer learning is done when a model is pre-trained on the same or a very similar modality to the final task to be solved. This paper demonstrates that transformers can be fine-tuned to comp...

#selfsupervisedlearning​ #yannlecun​ #facebookai​ Deep Learning systems can achieve remarkable, even super-human performance through supervised learning on large, labeled datasets. However, there are two problems: First, collecting ever more labeled data is expensive in both time and money. Second, these deep neural networks will be high performers on their task, but cannot easily generalize to ot...

#openai​ #clip​ #microscope​ OpenAI does a huge investigation into the inner workings of their recent CLIP model via faceted feature visualization and finds amazing things: Some neurons in the last layer respond to distinct concepts across multiple modalities, meaning they fire for photographs, drawings, and signs depicting the same concept, even when the images are vastly distinct. Through manual...

#machinelearning​ #phd​ #howto​ This video is advice for new PhD students in the field of Machine Learning in 2021 and after. The field has shifted dramatically in the last few years and navigating grad school can be very hard, especially when you're as clueless as I was when I started. The video is a personal recount of my mistakes and what I've learned from them. If you already have several publ...

In the recurring debate about bias in Machine Learning models, there is a growing argument saying that "the problem is not in the data", often citing the influence of various choices like loss functions or network architecture. In this video, we take a look at PAIR's AI Explorables through the lens of whether or not the bias problem is a data problem. OUTLINE: 0:00​ - Intro & Overview 1:45​ -...

#dreamcoder​ #programsynthesis​ #symbolicreasoning​ Classic Machine Learning struggles with few-shot generalization for tasks where humans can easily generalize from just a handful of examples, for example sorting a list of numbers. Humans do this by coming up with a short program, or algorithm, that explains the few data points in a compact way. DreamCoder emulates this by using neural guided sea...

#nerf​ #neuralrendering​ #deeplearning​ View Synthesis is a tricky problem, especially when only given a sparse set of images as an input. NeRF embeds an entire scene into the weights of a feedforward neural network, trained by backpropagation through a differential volume rendering procedure, and achieves state-of-the-art view synthesis. It includes directional dependence and is able to capture f...