van deventer
Distal Interference: Exploring the Limits of Model-Based Continual Learning
van Deventer, Heinrich, Bosman, Anna Sergeevna
Continual learning is the sequential learning of different tasks by a machine learning model. Continual learning is known to be hindered by catastrophic interference or forgetting, i.e. rapid unlearning of earlier learned tasks when new tasks are learned. Despite their practical success, artificial neural networks (ANNs) are prone to catastrophic interference. This study analyses how gradient descent and overlapping representations between distant input points lead to distal interference and catastrophic interference. Distal interference refers to the phenomenon where training a model on a subset of the domain leads to non-local changes on other subsets of the domain. This study shows that uniformly trainable models without distal interference must be exponentially large. A novel antisymmetric bounded exponential layer B-spline ANN architecture named ABEL-Spline is proposed that can approximate any continuous function, is uniformly trainable, has polynomial computational complexity, and provides some guarantees for distal interference. Experiments are presented to demonstrate the theoretical properties of ABEL-Splines. ABEL-Splines are also evaluated on benchmark regression problems. It is concluded that the weaker distal interference guarantees in ABEL-Splines are insufficient for model-only continual learning. It is conjectured that continual learning with polynomial complexity models requires augmentation of the training data or algorithm.
- Africa > South Africa > Gauteng > Pretoria (0.04)
- North America > United States > New York (0.04)
- North America > United States > Wisconsin > Dane County > Madison (0.04)
- (2 more...)
'Hey dude, do this': the last resort for female gamers escaping online abuse
In an extract from her book, Game Changers: From Minecraft to Misogyny, the Fight for the Future of Videogames (co-authored by Dan Golding), Leena van Deventer writes candidly about a time she was sexually harassed online. She was playing Team Fortress 2, an online multiplayer shooter, and one of her favourite games. Van Deventer had just "splurged" on a new headset with headphones and microphone with an exciting feature: voice modifiers "that made me sound like a cool robot or a huge giant". Once she began speaking in-game, however, there was what she calls "the reaction": "Omg are you like twelve, dude?" We got a girl in here, boys!" Van Deventer writes that she did her best to shrug off the comments and keep playing. One man, however, continued to "stalk" her avatar in the game. "Then came the questions: What are you wearing?