text
stringlengths 301
426
| source
stringclasses 3
values | __index_level_0__
int64 0
404k
|
|---|---|---|
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
you are a Beginner i suggest you to read A Beginner intro to Neural Networks and A Beginner intro to Convolutional Neural Networks. Now in this, we will learn: what are Neural Networks? what Neural Networks do? why not Neural Networks/Feed forward Networks? Why/what are Recurrent Neural Networks?
|
medium
| 8,176 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
Different Types of RNN’s Deep view into RNN’s Character level language model Back propogation through time(BTT) Issues of RNN’s? Advantages & Disadvantages of RNN Why LSTM’s? Forget gate input gate output gate. Resources What are Neural networks? Neural networks are set of algorithms inspired by
|
medium
| 8,177 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
the functioning of human brian. Generally when you open your eyes, what you see is called data and is processed by the Nuerons(data processing cells) in your brain, and recognises what is around you. That’s how similar the Neural Networks works. They takes a large set of data, process the
|
medium
| 8,178 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
data(draws out the patterns from data), and outputs what it is. What they do ? Neural networks sometimes called as Artificial Neural networks(ANN’s), because they are not natural like neurons in your brain. They artifically mimic the nature and funtioning of Neural network. ANN’s are composed of a
|
medium
| 8,179 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
large number of highly interconnected processing elements (neurones) working in unison to solve specific problems. ANNs, like people,like child, they even learn by example. An ANN is configured for a specific application, such as pattern recognition or data classification,Image recognition, voice
|
medium
| 8,180 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
recognition through a learning process. Neural networks (NN) are universal function approximaters so that means neural networks can learn an approximation of any function f() such that. y = f(x) NN trying to predict the image(data) that given to it. it predicts that the no is 2 here A Neural
|
medium
| 8,181 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
network/feed forward Neural network predicting cat image you can read more about Artificial Neural Networks here. Why Not Neural Networks/Feed forward Networks? A trained Feed Forward Neural Network can be Exposed to any huge random collection of images and asked to predict the output. for example
|
medium
| 8,182 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
check out the below figure In this training process, the first picture that the Neural network exposed to, will not necessarly alter how it classifies the Second one. here the output of Cat does not relate to the output Dog. There are several scenario’s where the previous understanding of data is
|
medium
| 8,183 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
important.for example: Reading book, understanding lyrics,..,. These networks do not have memory in order to understand Sequential data like Reading books. how do we overcome this challenge of understanding previous output? solution: RNN’s. What are RNN’s? The idea behind RNNs is to make use of
|
medium
| 8,184 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
sequential information. In a traditional neural network we assume that all inputs (and outputs) are independent of each other. But for many tasks that’s a very bad idea. If you want to predict the next word in a sentence you better know which words came before it. RNNs are called recurrent because
|
medium
| 8,185 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
they perform the same task for every element of a sequence, with the output being depended on the previous computations and you already know that they have a “memory” which captures information about what has been calculated so far. “Whenever there is a sequence of data and that temporal dynamics
|
medium
| 8,186 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
that connects the data is more important than the spatial content of each individual frame.” – Lex Fridman (MIT) More about RNN’s explained below. Different types of RNN’s The core reason that recurrent nets are more exciting is that they allow us to operate over sequences of vectors: Sequences in
|
medium
| 8,187 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
the input, the output, or in the most general case both. A few examples may make this more concrete: Different types of Recurrent Neural Networks. (2) Sequence output (e.g. image captioning takes an image and outputs a sentence of words). (3) Sequence input (e.g. sentiment analysis where a given
|
medium
| 8,188 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
sentence is classified as expressing positive or negative sentiment). (4) Sequence input and sequence output (e.g. Machine Translation: an RNN reads a sentence in English and then outputs a sentence in French). (5) Synced sequence input and output (e.g. video classification where we wish to label
|
medium
| 8,189 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
each frame of the video). Notice that in every case are no pre-specified constraints on the lengths sequences because the recurrent transformation (green) is fixed and can be applied as many times as we like. Each rectangle in above image represent Vectors and Arrows represent functions. Input
|
medium
| 8,190 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
vectors are Red in color, output vectors are blue and green holds RNN’s state. One-to-one: This also called as Plain/Vaniall Neural networks. It deals with Fixed size of input to Fixed size of Output where they are independent of previous information/output. Ex: Image classification. One-to-Many:
|
medium
| 8,191 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
it deals with fixed size of information as input that gives sequence of data as output. Ex:Image Captioning takes image as input and outputs a sentence of words. Deep Visual-Semantic Alignments for Generating Image Descriptions. Many-to-One: It takes Sequence of information as input and ouputs a
|
medium
| 8,192 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
fixed size of output. Ex:sentiment analysis where a given sentence is classified as expressing positive or negative sentiment. Many-to-Many: It takes a Sequence of information as input and process it recurrently outputs a Sequence of data. Ex: Machine Translation, where an RNN reads a sentence in
|
medium
| 8,193 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
English and then outputs a sentence in French. Bidirectional Many-to-Many: Synced sequence input and output. Notice that in every case are no pre-specified constraints on the lengths sequences because the recurrent transformation (green) is fixed and can be applied as many times as we like. Ex:
|
medium
| 8,194 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
video classification where we wish to label each frame of the video. CNN vs RNN: I dont think you need explanation for this. You can easily get what it is just by looking at the figure below: Difference in Describing images using CNN , RNN. Deep view into RNN’s: In a simple Neural Network you can
|
medium
| 8,195 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
see Input unit, hidden units and output units that process information independently having no relation to previous one. Also here we gave different weights and bias to the hidden units giving no chance to memorize any information. A single Neuron From NN vs RNN’s Source:Geeks for Geeks where
|
medium
| 8,196 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
Hidden layer in RNN’s have same weights and bias through out the process giving them the chance to memorize information processed through them. Current time stamp: look at the above figure, where the formula for the Current state: function using previous times stamp and input gives current time
|
medium
| 8,197 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
stamp where Ot is output state, ht →current time stamp, ht-1 → is previous time stamp, and xt is passed as input state. Applying activation function: Activation function W is weight, h is the single hidden vector, Whh is the weight at previous hidden state, Whx is the weight at current input state.
|
medium
| 8,198 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
Where tanh is the Activation funtion, that implements a Non-linearity that squashes the activations to the range[-1.1] Output: output Yt is the output state. Why is the weight at the output state. Example: “Character level language model” explained below. Character level language model: We’ll give
|
medium
| 8,199 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
the RNN a huge chunk of text and ask it to model the probability distribution of the next character in the sequence given a sequence of previous characters. As a working example, suppose we only had a vocabulary of four possible letters “helo”, and wanted to train an RNN on the training sequence
|
medium
| 8,200 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
“hello”. This training sequence is in fact a source of 4 separate training examples: 1. The probability of “e” should be likely given the context of “h”, 2. “l” should be likely in the context of “he”, 3. “l” should also be likely given the context of “hel”, and finally 4. “o” should be likely
|
medium
| 8,201 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
given the context of “hell”. you can get more about this example here and here. Backpropogate Through Time: To understand and visualize the Backpropogation, lets unroll the network at all the time stamps, so that you can see how the weights gets updated.Going back in Every time stamp to
|
medium
| 8,202 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
change/update the weights is called Backpropogate through time. We typically treat the full sequence (word) as one training example, so the total error is just the sum of the errors at each time step (character). The weights as we can see are the same at each time step. Let’s summarize the steps
|
medium
| 8,203 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
for backpropagation through time The cross entropy error is first computed using the current output and the actual output Remember that the network is unrolled for all the time steps For the unrolled network, the gradient is calculated for each time step with respect to the weight parameter Now
|
medium
| 8,204 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
that the weight is the same for all the time steps the gradients can be combined together for all time steps The weights are then updated for both recurrent neuron and the dense layers Note: Going back into every time stamp and updating its weights is really a slow process. It takes both the
|
medium
| 8,205 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
computational power and time. While Backpropogating you may get 2 types of issues. Vanishing Gradient Exploding Gradient Vanishing Gradient: where the contribution from the earlier steps becomes insignificant in the gradient descent step. While you are using Backpropogating through time, you find
|
medium
| 8,206 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
Error is the difference of Actual and Predicted model. Now what if the partial derivation of error with respect to weight is very less than 1? If the partial derivation of Error is less than 1, then when it get multiplied with the Learning rate which is also very less. then Multiplying learning
|
medium
| 8,207 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
rate with partial derivation of Error wont be a big change when compared with previous iteration. For ex:- Lets say the value decreased like 0.863 →0.532 →0.356 →0.192 →0.117 →0.086 →0.023 →0.019.. you can see that there is no much change in last 3 iterations. This Vanishing of Gradience is called
|
medium
| 8,208 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
Vanishing Gradience. Vanishing Gradient: where the contribution from the earlier steps becomes insignificant in the gradient for the vanilla RNN unit. Aslo this Vanishing gradient problem results in long-term dependencies being ignored during training. you Can Visualize this Vanishing gradient
|
medium
| 8,209 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
problem at real time here. Several solutions to the vanishing gradient problem have been proposed over the years. The most popular are the aforementioned LSTM and GRU units, but this is still an area of active research. Exploding Gradient: We speak of Exploding Gradients when the algorithm assigns
|
medium
| 8,210 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
a stupidly high importance to the weights, without much reason. But fortunately, this problem can be easily solved if you truncate or squash the gradients. Exploding Gradient similarly here, What if the Partial derivation of Errror is more than 1? Think. How can you overcome the Challenges of
|
medium
| 8,211 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
Vanishing and Exploding Gradience? Vanishing Gradience can be overcome with Relu activation function. LSTM, GRU. 2. Exploding Gradience can be overcome with Truncated BTT(instead starting backprop at the last time stamp, we can choose similar time stamp, which is just before it.) Clip Gradience to
|
medium
| 8,212 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
threshold. RMSprop to adjust learning rate. Advantages of Recurrent Neural Network The main advantage of RNN over ANN is that RNN can model sequence of data (i.e. time series) so that each sample can be assumed to be dependent on previous ones Recurrent neural network are even used with
|
medium
| 8,213 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
convolutional layers to extend the effective pixel neighborhood. Disadvantages of Recurrent Neural Network Gradient vanishing and exploding problems. Training an RNN is a very difficult task. It cannot process very long sequences if using tanh or relu as an activation function. Long Short Term
|
medium
| 8,214 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
Memory: Long Short Term Memory with its gates A special kind of RNN’s, capable of Learning Long-term dependencies. LSTM’s have a Nature of Remembering information for a long periods of time is their Default behaviour. The repeating module in LSTM with its gates. LSTM had a three step Process: look
|
medium
| 8,215 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
at the below figure that says Every LSTM module will have 3 gates named as Forget gate, Input gate, Output gate. LSTM with its Gates Forget Gate: Decides how much of the past you should remember. This gate Decides which information to be omitted in from the cell in that particular time stamp. It is
|
medium
| 8,216 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
decided by the sigmoid function. it looks at the previous state(ht-1) and the content input(Xt) and outputs a number between 0(omit this)and 1(keep this)for each number in the cell state Ct−1. Forwad Gate EX: lets say ht-1 →Roufa and Manoj plays well in basket ball. Xt →Manoj is really good at
|
medium
| 8,217 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
webdesigning. Forget gate realizes that there might be change in the context after encounter its first fullstop. Compare with Current Input Xt. Its important to know that next sentence, talks about Manoj. so information about Roufa is omited. Update Gate/input gate: Decides how much of this unit is
|
medium
| 8,218 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
added to the current state. Input/Update gate Sigmoid function decides which values to let through 0,1. and tanh function gives weightage to the values which are passed deciding their level of importance ranging from-1 to 1. EX: Manoj good webdesigining, yesterday he told me that he is a university
|
medium
| 8,219 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
topper. input gate analysis the important information. Manoj good webdesigining, he is university topper is important. yesterday he told me that is not important, hence forgotten. Output Gate: Decides which part of the current cell makes it to the output. Output Gate Sigmoid function decides which
|
medium
| 8,220 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
values to let through 0,1. and tanh function gives weightage to the values which are passed deciding their level of importance ranging from-1 to 1 and multiplied with output of Sigmoid. EX: Manoj good webdesigining, he is university topper so the Merit student _______________ was awarded University
|
medium
| 8,221 |
Machine Learning, Deep Learning, Data Science, Artificial Intelligence, Python.
Gold medalist. there could be lot of choices for the empty dash. this final gate replaces it with Manoj. A Blog on LSTM’s with Nice visualization is here. Resources: Distill website and github . Chris olah blog here . More on Andrej karpathy blog here . More on Visualizing Memorization in RNN’s .
|
medium
| 8,222 |
.
has grown increasingly cranky, tired, and creatively bereft, it’s because, in our text- and language-obsessed era, we are. We have created a nation of citizens whose survival brains are on overdrive and whose creative brains have effectively been relegated to the corner. Change starts with flipping
|
medium
| 8,231 |
.
polarization, establishing shared values, giving everyone a seat at the table, and allowing everyone to be their best creative self. It is precisely this hands-on approach that could be taken to address polarization within the rancorous school board meetings, townhalls, and city council meetings of
|
medium
| 8,241 |
.
explore the world — collaboration simply unfolds. We also at no point insist that people use certain words or certain language. In the same way that we place no premium on perfection in people’s models, we don’t expect perfection in people’s language. Particularly amongst highly educated circles, a
|
medium
| 8,245 |
SEO, Blog, Blogging, Keyword Research, Keywords.
Free 1-Click Keyword Clustering Tool for SEO Unlocking the Power of Keyword Clustering for Your SEO Success In this article, I will tell you how you can create your own keyword cluster, easily and for free, by using the tool named ClusterKeywords.com. Before we start, I will give you a small
|
medium
| 8,256 |
SEO, Blog, Blogging, Keyword Research, Keywords.
overview. Keyword clusters are basically related word groups, as you can see in this diagram. If your topic is screen care, the related articles will be Screen Care Combination, Screen Oily, Screen Dry, Screen Care Par Acne, and How to Get Red Off Wrinkles on Screen. These keywords are all related
|
medium
| 8,257 |
SEO, Blog, Blogging, Keyword Research, Keywords.
and can be grouped together in a cluster. We strongly recommend that you check out our guide on how to take advantage of AI in today’s passive income economy. The Power of Keyword Clustering The most related keywords, like screen care, screen oily, screen dry, etc., can be found within a single
|
medium
| 8,258 |
SEO, Blog, Blogging, Keyword Research, Keywords.
cluster because they are very much related to each other. If you research these keywords and divide them into clusters, it becomes much easier to create content that targets the right audience and improves your SEO. To find keywords for your website or blog, you can head over to
|
medium
| 8,259 |
SEO, Blog, Blogging, Keyword Research, Keywords.
ClusterKeywords.com, a free tool that makes keyword clustering a breeze. Simply enter your primary topic, such as “leather jacket,” and the tool will generate a list of related keywords and group them into clusters for you. Leveraging Keyword Clusters for Effective Content Creation Once you have
|
medium
| 8,260 |
SEO, Blog, Blogging, Keyword Research, Keywords.
your keyword clusters, you can use them to create targeted content for your website or blog. For example, if you have a website on leather jackets, you can create different categories based on the clusters, such as “Bike and Jacket,” “Women’s Leather,” and “Leather Jacket Outfits.” By writing
|
medium
| 8,261 |
SEO, Blog, Blogging, Keyword Research, Keywords.
content that covers the keywords within each cluster, you’re telling Google that you’re an authority on the topic. This can help you rank higher in search results and drive more organic traffic to your website. Analyzing Keyword Competition When creating content based on your keyword clusters, it’s
|
medium
| 8,262 |
SEO, Blog, Blogging, Keyword Research, Keywords.
important to do a little due diligence on the individual keywords. Check the search volume, keyword difficulty, and the top-ranking content for each keyword to see how competitive the landscape is. For example, if you find a keyword like “Over Size Leather Jacket Outfit” with a decent search volume
|
medium
| 8,263 |
SEO, Blog, Blogging, Keyword Research, Keywords.
and low competition, you can create a targeted article around that keyword and include related terms from the cluster. This will help you build a strong, topical content strategy that can boost your SEO performance. Implementing Keyword Clustering in Your Content Strategy To give you an example of
|
medium
| 8,264 |
SEO, Blog, Blogging, Keyword Research, Keywords.
how I would implement keyword clustering, let’s say I have a website on dinosaurs and the omniverse. I’ve created a 3,000-word article on this topic, and within the article, I’ve included internal links to other relevant content on my website, such as articles on individual dinosaur species. By
|
medium
| 8,265 |
SEO, Blog, Blogging, Keyword Research, Keywords.
linking these related articles back to the pillar page (the 3,000-word article), I’m creating a tight, topical cluster that shows Google I’m an authority on this subject. This approach can be applied to any niche or industry, and the more you can interlink your content within these keyword
|
medium
| 8,266 |
SEO, Blog, Blogging, Keyword Research, Keywords.
clusters, the stronger your overall SEO strategy will be. Remember, the key to successful keyword clustering is to research your topics thoroughly, analyze the competition, and create high-quality, targeted content that serves your audience. With the help of a free tool like ClusterKeywords.com,
|
medium
| 8,267 |
SEO, Blog, Blogging, Keyword Research, Keywords.
you can streamline this process and unlock new SEO opportunities for your business. We strongly recommend that you check out our guide on how to take advantage of AI in today’s passive income economy. Cubed Thank you for reading until the end. Before you go: Please consider clapping and following
|
medium
| 8,268 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
System Change Portfolios for Circular Economy — From Design to Activation Early Musings as we Engage in the Next Frontier of Our Work by Alex Oprunenco, Brent Wellsch and Tommaso Balbo di Vinaldo It is widely recognized that our world is facing unprecedented uncertainty driven by the compounded
|
medium
| 8,270 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
effects of novel and familiar crises. Dealing with those requires us shifting from technical “quick fixes” to approaches that acknowledge and embrace complexity. This is a journey that we embarked on with several city administrations and partners in Asia with an ambition to achieve a “triple ring”
|
medium
| 8,271 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
of coherence for decision-making: policy coherence — ensuring proposed “interventions” are coherent with the challenge at hand and amongst themselves ecosystem and partnership coherence — it is co-owned and can be sustainably managed long-term by community actors and relevant stakeholders.
|
medium
| 8,272 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
resourcing coherence (more on this soon) — investments from various sources conform with an overall portfolio logic and are reinforcing rather than piecemeal. Our “Sense and Frame; Engage and Position; and Transform” framework was a “learning capture”. Now as we move into portfolio activation we
|
medium
| 8,273 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
are discovering what a shift from projects to portfolio entails and how it can support better decision-making and resource alignment. While we seek to affect change externally, we know we must evolve ourselves, as individuals and organizations. The following challenges in practice have surfaced as
|
medium
| 8,274 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
our system change portfolio for circular economy in Pasig City, Philippines is being activated: Managing a Cultural Shift Changing our Decision-Making Calculus; The Need for A Different Monitoring, Evaluation, and Learning Methods (MEL) to Support Decision-making in complexity Portfolio Activation
|
medium
| 8,275 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
and Challenges in Practice Managing a Cultural Shift Harnessing and maintaining the mindset required to activate a system change portfolio, and then keeping it dynamic and alive is crucial. The portfolio is a vessel for system learning and discovery, but the temptation to become myopic or obsessed
|
medium
| 8,276 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
with a specific initiative or intervention in the portfolio (or even worse to assume you have figured out what works with certainty) is ever present. However, it’s one thing to know that we need this mindset (or write about it in a blog 🙂) and it’s another thing to adopt it in practice. Drucker’s
|
medium
| 8,277 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
“culture eats strategy for breakfast” principle is very relevant here. This way of thinking and working is in many ways counter-cultural to how large bureaucratic organizations have traditionally operated. Acknowledging this need to work with and not against current systemic realities, we have
|
medium
| 8,278 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
honed in on Meadow’s advice to dance with the systems we are engaging with through what we are calling a demonstration portfolio. The purpose of this demonstration portfolio is to allow us to walk before we run through an initial portfolio that consists of complementary and aligned interventions
|
medium
| 8,279 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
that can foster experimentation and learning about the challenge space itself. In Pasig City we have developed a demonstration portfolio that will align initial interventions across three key thematic areas: nurturing new markets, participatory involvement of consumers and citizens, and building an
|
medium
| 8,280 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
entrepreneurial city government. Through the demonstration portfolio we hope to: Externally learn about what aspects of our portfolio works, what does not, what needs to be discarded, built in, and so on. Internally learn about what it takes (skill-sets, team make-up and competencies, processes,
|
medium
| 8,281 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
etc.) to successfully manage a system change portfolio. It may seem simple, but we are hoping that the activation of a demonstration portfolio will help keep us humble and open to emergent learnings that will benefit our ambitions in the long-term. Our work in Pasig City is our first attempt in
|
medium
| 8,282 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
doing so and we will look to apply this experimental approach in two other contexts over the course of 2023. Changing our Decision-Making Calculus: Again, referencing the Pasig City example, we are seeking to find the fine balance between striving for what is ideal while being open to the optimal
|
medium
| 8,283 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
possibility of right now. This kind of work runs the risk of the infamous “paralysis by analysis” where insecurity drives a need to constantly gather more data/evidence which becomes burdensome and can result in a limbo state of inaction. In our work we have come up against this dynamic in a very
|
medium
| 8,284 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
real way. Reconciling pre-determined funding decisions for things like infrastructure upgrades to support improvements in waste management processes and practices is one example. Our real-world reality was that funding to support infrastructure upgrades had already been secured and questioning
|
medium
| 8,285 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
whether this was the right funding for the right infrastructure was not a possibility. As such, our focus here has been spent not disputing this reality but reasoning how to best support infrastructure upgrades as part of a larger portfolio logic and structure that ensure that these upgrades are
|
medium
| 8,286 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
complimented with other endeavors (e.g. skill-training) to ensure that this equipment is used for its intended purpose and is set-up for success. As this simple example shows we are trying to become comfortable with a decision-making process that is: Experimental and action biased; Pragmatic in
|
medium
| 8,287 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
balancing what is ideal to do versus what we can do right now; Premised on learning So what does it mean for resource allocation? While in the very early stages, we have started to explore how portfolios may affect decision-making processes related to financial resource allocation. At this point we
|
medium
| 8,288 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
have reasoned that a portfolio approach has the capacity to direct this effort in two ways: Directionality — we have observed/experienced many circumstances where financial resource allocation decisions are informed by a deficit mindset, and yet in the end some available funding goes unspent or is
|
medium
| 8,289 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
not spent effectively (e.g. many cities in the Philippines underspent available budgets). A portfolio approach can help this drawback by providing direction and decision-making roadmap for how available public resources should be allocated. For example, the Pasig City portfolio has influenced how
|
medium
| 8,290 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
city administration thinks about an SME incentive program by providing a logic for how this investment can have an impact beyond the program itself (see our city government pivot); Alignment — a portfolio’s narrative and content not only gives a different “north star” for the money, it creates a
|
medium
| 8,291 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
playing field for alignment, where different types of instruments can combine for system-level impact in an articulated and logical way (from citizen and conversion grants to infrastructure co-investment and grant-equity funds). This alignment brings clarity to what private, public, donors, and
|
medium
| 8,292 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
venture funds can achieve when channeled in unison and at what stage–opposite to what we may usually observe now when public and private funders pursue their own institutional agendas. A Different Monitoring, Evaluation, and Learning Methods (MEL) to support decision-making in complexity If our
|
medium
| 8,293 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
intent in developing and implementing portfolios to support decision-making and resource allocation to deal with novel complex challenging holds then our bet is that learning has to play a much more strategic role than it does now, i.e. not an “add-on” or a shortcut to reporting on results. Indeed,
|
medium
| 8,294 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
as this work is highly contextual, adaptive and relational, continuous learning is the only way you can engage with the system and observe your portfolio does no harm. In our process we are seeking to embed learning into the core of our monitoring and evaluation (M&E) efforts. In doing so, we are
|
medium
| 8,295 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
co-opting what can be a very linear and stagnant methodological field to be dynamic and reflective. Many M&E methods are biased towards linear approaches and have long been criticized for their methodological stagnation — a large majority of organizations still use traditional M&E methodologies
|
medium
| 8,296 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
that, even if they are solid and well developed, fail often to capture the complexity they are trying to measure. However, system change is indeed not linear and the non-linear reality of this work inclines us to put more emphasis on measuring the contribution of our effort, rather than
|
medium
| 8,297 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
attribution. By the same token, measuring systems requires teams to firstly define and ‘see’ the system by identifying the dynamic attributes of the system: it is not possible to assess whether or how systemic change has occurred. M&E methods need to be adapted and revised when considering and
|
medium
| 8,298 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
measuring ‘system change’. Two ways we are initially attempting to do so are, as follows: 1. Mixing agile with M&E qualitative approaches with a strong focus on learning. In an attempt of being pragmatic and able to measure impact, we have developed some minimum guidelines for what is required from
|
medium
| 8,299 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
a M&E perspective to launch the demonstration portfolio in Pasig City. We reasoned we need to: Clearly define the outcomes desired within the demonstration portfolio along with progress indicators unique to specific interventions and the portfolio (methodology inspired by outcome mapping);
|
medium
| 8,300 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
Establish a narrative for change but also a concrete plan of action of how the team wants to impact change with a set of clearly established responsibilities; Integrate learning mechanisms into all activities through a 3 step process that embraces elements of SCRUM methodology (check-in, review and
|
medium
| 8,301 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
retrospective) — see the diagram below 2. Adding the system perspective and integrating it into the framework/templates As the core objective is to manage and implement a portfolio of interventions and to learn from it, we are adding a system perspective into each M&E activity we are developing.
|
medium
| 8,302 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
For instance, when doing weekly check-ins, we are suggesting that the portfolio management teams pay attention (‘intentional learning’) not only to the activities they have implemented and the challenges they are facing but also to the emerging interconnections between interventions and the
|
medium
| 8,303 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
‘unexpected’ effects that those interventions are having on the system as we have defined (see the template below.) We have worked diligently to pre-define what these interconnections might be but need to be alive to what will actually happen through practical application. Again, these are some
|
medium
| 8,304 |
System Change, Portfolio Approach, The Circular Economy, Portfolio Activation.
initial challenges we have surfaced and will keep track of as we progress. We look forward to working out loud and sharing updated findings on these, as well as other phenomena as it emerges. Stay tuned 😊 We are extremely privileged to work with Alyanna, Francis “Kapi”, Gwen, Folay, and Rex (UNDP
|
medium
| 8,305 |
Machine Learning, Data Science, Synthetic Data, Data Architecture, Data.
Storing vast amounts of data comes with its own risks and challenges. Synthetic data is one of the options in the toolkit to address them. This article reviews different angles, including data confidentiality, retention and de-identification. Census data is unreliable. There are several reasons;
|
medium
| 8,307 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.