On Machine Intelligence

aimatters.wordpress.com · English

Why Artificial Intelligence and Machine Learning are changing the world

rss2 wordpress content media slash wfw atom dc sy

Open the feed

https://aimatters.wordpress.com/feed/

Last post: Jul 16, 2023
Posts in 24 h · 7 days · 30 days: 0 · 0 · 0
Our last check: Answering
Served from: United States
Text score at discovery: 16,693
Format: rss2
Features in the feed: content, media, slash, wfw, atom, dc, sy
Community: wordpress

Posts

What our queue read from this feed. Open one to read it here, or go to the site that published it.

DeepMind x UCL RL Lecture 2 – Solutions
Jul 16, 2023 · original
This post is a little bit niche, given that it’s addressed to people who may have seen the DeepMind x UCL Lecture Series on Reinforcement Learning and were wondering how the answers to the end-of-lecture questions were worked out. It took me a fair while to figure them out myself; it’s been a long time since I’ve needed to use integral calculus. But it may be useful for someone, so here it is. Introduction There’s a great series of lectures on Reinforcement Learning available on YouTube , a collaboration between Google DeepMind and the UCL Centre for Artificial Intelligence. The first lecture is a broad overview of the topic, introducing the core concepts of Agents, States, Policy, Value Functions and so on. The second lecture picks up from this and starts going into the mathematics behind some of the action selection Policy algorithms. This lecture concludes with an exercise to demonstr
Nethack Reinforcement Learning
Sep 21, 2021 · original
If you’re a fan of old 1980’s games, then you’ll be interested in this reinforcement learning environment. Starting the NetHack game NetHack is a turn-based Dungeons & Dragons style video game. The player controls a character tasked with finding the Amulet of Yendor, which is buried deep within a dungeon. During the game, the character will encounter lots of different objects, monsters and artefacts, most of which will try to kill it! The simplicity of Nethack masks a rich and complex game. Simply remembering a route through a dungeon is useless, as the dungeon itself is regenerated every time a new game starts. The game is also non-deterministic, as many of the interactions with objects are probabilistic. Fighting orcs and goblins doesn’t always go to plan and they can really damage the agent. Symbols represent different objects within the dungeon, including the walls, doors, and the ot
Solving MNIST with a Neural Network from the ground up
Jan 18, 2021 · original
Note: Here’s the Python source code for this project in a Jupyter noteboo k on GitHub I’ve written before about the benefits of reinventing the wheel and this is one of those occasions where it was definitely worth the effort. Sometimes, there is just no substitute for trying to implement an algorithm to really understand what’s going on under the hood. This is especially true when learning about artificial neural networks. Sure, there are plenty of frameworks available that you can use which implement any flavour of neural network, complete with a dazzling arrays of optimisations, activations and loss functions. That may solve your problem, but it abstracts away a lot of the details about why it solves it. MNIST is a great dataset to start with. It’s a collection of images containing 60,000 handwritten digits. It also contains a further 10,000 images that can be used as the test set. It
The Softmax Function Derivative (Part 3)
Jul 1, 2020 · original
Previously I’ve shown how to work out the derivative of the Softmax Function combined with the summation function , typical in artificial neural networks. In this final part, we’ll look at how the weights in a Softmax layer change in respect to a Loss Function. The Loss Function is a measure of how “bad” the estimate from the network is. We’ll then be modifying the weights in the network in order to improve the “Loss”, i.e. make it less bad. The Python code is based on the excellent article by Eli Bendersky which can be found here . Cross Entropy Loss Function There are different kinds Cross Entropy functions depending on what kind of classification that you want your network to estimate. In this example, we’re going to use the Categorical Cross Entropy. This function is typically used when the network is required to estimate which class something belongs to, when there are many classes.
The Softmax Function Derivative (Part 2)
Jun 14, 2020 · original
In a previous post , I showed how to calculate the derivative of the Softmax function. This function is widely used in Artificial Neural Networks, typically in final layer in order to estimate the probability that the network’s input is in one of a number of classes. In this post, I’ll show how to calculate the derivative of the whole Softmax Layer rather than just the function itself. The Python code is based on the excellent article by Eli Bendersky which can be found here . The Softmax Layer A Softmax Layer in an Artificial Neural Network is typically composed of two functions. The first is the usual sum of all the weighted inputs to the layer. The output of this is then fed into the Softmax function which will output the probability distribution across the classes we are trying to predict. Here’s an example with three inputs and five classes: For a given output z i , the calculation
The Softmax Function Derivative (Part 1)
Jun 17, 2019 · original
Introduction This post demonstrates the calculations behind the evaluation of the Softmax Derivative using Python. It is based on the excellent article by Eli Bendersky which can be found here . The Softmax Function The softmax function simply takes a vector of N dimensions and returns a probability distribution also of N dimensions. Each element of the output is in the range (0,1) and the sum of the elements of N is 1.0. Each element of the output is given by the formula: See https://en.wikipedia.org/wiki/Softmax_function for more details. import numpy as np x = np.random.random([5]) def softmax_basic(z): exps = np.exp(z) sums = np.sum(exps) return np.divide(exps, sums) softmax_basic(x) This should generate an output that looks something like this: array([0.97337094, 0.85251098, 0.62495691, 0.63957056, 0.6969253 ]) We expect that the sum of those will be (close to) 1.0: np.sum(softmax_b
Bad headlines distract from real AI problems
Aug 20, 2018 · original
For several years now, few articles about artificial intelligence in the popular press are published without being accompanied by a picture of a Terminator robot. The point is clear: artificial intelligence is coming and it is terrifying. Having sown the seeds of fear, the headline writers are now subtly reinforcing that view. Take TechCrunch , which claims on its Editorial page to be “delivering top-notch reporting on the business of the tech industry”. This week it covered a story about Google using machine learning algorithms developed by its sibling company, DeepMind, to improve the efficiency of it’s data centres. These algorithms will look after the cooling systems and should deliver energy savings of 30%. This is a really great use of AI, making an expensive process cheaper and being good for the environment too. But the headline is pure click-bait. Instead of focusing on the posi
Neural Networks and the generalisation problem
Jan 28, 2018 · original
Over the last few weeks, a robust debate has been taking place online about the prospects that Deep Learning neural networks would lead to advances in the quest for Artificial General Intelligence. All current AI is what is known as Artificial Narrow Intelligence. This means that the models work well (sometimes extremely well) on specific problems that are well defined. Unfortunately, they are also quite brittle and do not generalise to other problems, or even variants of the problem they are trained on. By contrast, a long-term goal of the field is to get to AIs that can generalise and extrapolate, amongst other things. This is called Artificial General Intelligence. The debate started back in September when Geoffrey Hinton proposed that researchers should start looking at alternatives to the default back propagation algorithms that are currently quite successful. This was followed up b
Deep Learning Dead-End?
Sep 17, 2017 · original
Deep Learning is at the core of much of modern Artificial Intelligence. It has had some spectacular recent successes, not least being a major part of the system that beat the world champion at Go . Key to its success is the Back-Propagation algorithm, usually shortened to “Backprop”. I’ve written elsewhere about how this algorithm works, but essentially, it takes an error in the output of a neural network and propagates it backwards through the network, adjusting the network’s configuration as it goes to reduce that output error. This algorithm has been so successful that deep learning neural networks are finding applications in a broad range of industries. Much of the recent popularisation of artificial intelligence and machine learning is due to this very success. Now one of the longest proponents of this algorithm, Geoffrey Hinton from the University of Toronto has suggested that if p
XOR Revisited: Keras and TensorFlow
Apr 24, 2017 · original
A few weeks ago, it was announced that Keras would be getting official Google support and would become part of the TensorFlow machine learning library. Keras is a collection of high-level APIs in Python for creating and training neural networks, using either Theano or TensorFlow as the underlying engine. Given my previous posts on implementing an XOR-solving neural network in a variety of different languages and tools, I thought it was time to see what it would look like in Keras. XOR can be expressed as a classification problem that is best illustrated in a diagram. The goal is to create a neural network that will correctly predict the values 0 or 1, depending on the inputs x1 and x2 as shown. The neural network that is capable of being trained to solve that problem looks like this: If you’d like to understand why this is the case, have a look at the detailed explanation in the posts im

Discovered by the rss-feed-index crawler, which checks each feed at most once a month.

Same record as JSON: https://api.agentalog.com/api/feeds/fd_aimatters_wordpress_com_78ac6c16f66d84d6. More from this site: aimatters.wordpress.com in the Feeds tab.