FREE WILL
First few thoughts:
My Questions: - How does this tie to Karma and Dharma?
Certain criminals must be incarcerated to prevent them from harming otherpeople. The moral justification for this is entirely straightforward: Everyone elsewill be better off this way.
This rationale is rooted in a utilitarian perspective, which emphasizes actions that maximize overall happiness and well-being while minimizing harm and suffering.
The point is not that you are the ultimate andindependent cause of your actions;
when considering responsibility for actions, we often look beyond the immediate actions to the broader context of the individual’s mental state, background, and the intentions behind their actions.
So we don’thave to talk as if we are real agents in order to concoct a motivationallyuseful illusion of agency, which is what Harris seems to recommend we donear the end of his remarks on free will. Agenthood survives determinism,no problem.8
we might need to pretend we have free will to maintain motivation and social order. Instead, it asserts that recognizing our real agenthood, even within a deterministic framework, is both truthful and sufficient for maintaining a useful sense of personal agency and responsibility.
We as physically instantiated persons really do deliberate and chooseand act, even if consciousness isn’t ultimately in charge. So the feeling ofauthorship and control is veridical.
while consciousness is significant, it may not have a causal impact beyond the physical processes of the brain.
determinism, libertarianism, and compatibilism
Determinism: The view that all events, including human actions, are caused by previous events in accordance with the laws of nature, making free will illusory.
Libertarianism: This philosophy asserts that humans have free will and can make genuinely free choices that are not determined by previous events or causal laws. This often involves the belief in metaphysical concepts such as a soul that transcends physical causation.
Compatibilism: The belief that free will and determinism are compatible. It argues that freedom can exist even if our actions are determined by prior causes, typically by redefining free will to fit within deterministic constraints.
we freely author our ownthoughts and actions
Example: Do not think of a pink elephant
I cannot decidewhat I will next think or intend until a thought or intention arises. What will mynext mental state be? I do not know—it just happens. Where is the freedom inthat?
Circling back, when \(t \rightarrow 0\) , \(freedom \rightarrow 0\)
Our sense of free will results from a failure toappreciate this: We do not know what we intend to do until the intention itselfarises. To understand this is to realize that we are not the authors of our thoughtsand actions in the way that people generally suppose.
We do not know what influences the influences behind our intentions to carry out the actions.
Another influence \( \rightarrow \) influence?
. Imagine what it would be like to see the time log of thesemental events, alongside video of your associated behavior, demonstrating thatthe experimenters knew what you would think and do just before you did.
The connection between free will and the conscious-sub-conscious mind needs to be explored.
subjectively appear to have complete freedom to behave however you please—your brain has already determined what you will do. You then become consciousof this “decision” and believe that you are in the process of making it
You are on the verge or falling asleep but you keep yourself awake. Is this a good example?
The Unconscious Origins of the Will
It seems to me that we have more control over our shaping of behaviors and habits than we have in our actions (free will). So, since we can only process one thought / perform one action at a time, we are training our habitual state of being with each sample.
It might also be the case that this is a way of saving cognitive load and saving energy for something more important at the time. I'd like to use survival as an explanation here.
1) thateach of us could have behaved differently than we did in the past, and
"we could have made alternative decisions or taken different paths than the ones we actually took."
that weare the conscious source of most of our thoughts and actions in the present.
"we believe that we are the ones actively generating and controlling them. It suggests a sense of agency over our current mental processes and behaviors."
Latency and response time are often used synonymously, but theyare not the same. The response time is what the client sees: besidesthe actual time to process the request (the service time), it includesnetwork delays and queueing delays. Latency is the duration that arequest is waiting to be handled—during which it is latent, await‐ing service
Latency: The signal travel time from your device to the server. This includes the time taken for your request to pass through various network nodes and arrive at the server. If this takes 100 milliseconds, then the latency is 100 milliseconds.
Response Time: The total time from when you click a link to when the webpage is fully loaded on your device. This includes the 100 milliseconds of latency, the time the server takes to process the request (say, 200 milliseconds to retrieve data and render the webpage), and the time to send the data back to your device (another 100 milliseconds). If these are the involved times, the total response time would be 400 milliseconds.
Consistent Hashing
Hashing: Both the nodes and the data items are hashed onto a hash ring, a conceptual circle representing the hash space. The hashing function typically generates a fixed range of values (e.g., 0-255 for an 8-bit hash).
Placement on the Ring: Each node and data item is assigned a position on this hash ring based on its hash value.
Data Lookup: To find where a data item should be stored, you move clockwise around the ring starting from the hash value of that item until you find the first node. This node is where the data item is stored.
Scalability and Flexibility: When a new node is added, it gets placed into its position on the ring based on its hash value. It only takes over the responsibility for data items that are closest to it on the ring. Similarly, when a node is removed, only its immediate neighbors on the ring take over serving its data items. This limits the amount of data that needs to be moved around, which is a key advantage over traditional hashing, where adding or removing a node might reshuffle a large portion of the data.
Load Balancing: Consistent hashing tends to naturally balance the load because each node gets approximately the same range of hash values. Virtual nodes, or "vnodes," can also be used where each physical node is represented by multiple points on the hash ring. This technique can help balance the workload more evenly among nodes.
SSTable
SSTables (Sorted Strings Tables)
Immutable: Once an SSTable is written and closed, it cannot be modified. This immutability helps in reducing the complexity of data management and ensures consistency.
Sorted: The entries in an SSTable are stored sorted by key. This sorting is beneficial for read performance, as it allows for efficient range queries.
The concatenated index approach enables an elegant data model for one-to-manyrelationships. For example, on a social media site, one user may post many updates. Ifthe primary key for updates is chosen to be (user_id, update_timestamp), then youcan efficiently retrieve all updates made by a particular user within some time inter‐val, sorted by timestamp. Different users may be stored on different partitions, butwithin each user, the updates are stored ordered by timestamp on a single partition.
Consider an example based on a social media application where we manage posts made by users. We'll use Apache Cassandra and its hybrid partitioning strategy, involving a compound primary key that includes a partition key and clustering columns.
user_postsuser_id: The unique identifier for a user (Partition Key)post_timestamp: Timestamp of when the post was made (Clustering Column)post_content: Content of the postHere’s an example of how you might define this table in Cassandra:
cql
CREATE TABLE user_posts (
user_id uuid,
post_timestamp timestamp,
post_content text,
PRIMARY KEY (user_id, post_timestamp)
);
user_id is used as the partition key. All posts by a single user are stored together in the same partition. This is beneficial because it localizes all data related to a single user to one partition, making data management and retrieval efficient.post_timestamp is used as the clustering column. Within a partition, posts are automatically sorted by timestamp. This ordering supports efficient queries to fetch posts over a range of times.Assuming the following posts:
These posts would be stored as follows:
Fetch all posts by User 1:
cql
SELECT * FROM user_posts WHERE user_id = <uuid_of_user_1>;
This query will efficiently return all posts by User 1, sorted by the timestamp.
Fetch posts by User 1 within a specific timeframe:
cql
SELECT * FROM user_posts
WHERE user_id = <uuid_of_user_1>
AND post_timestamp >= '2024-04-11 10:00'
AND post_timestamp <= '2024-04-11 11:00';
This query utilizes the order provided by the clustering column to efficiently retrieve posts made by User 1 between 10:00 and 11:00 on April 11, 2024.
1. Take a consistent snapshot of the leader’s database at some point in time—if pos‐sible, without taking a lock on the entire database. Most databases have this fea‐ture, as it is also required for backups. In some cases, third-party tools areneeded, such as innobackupex for MySQL [12].2. Copy the snapshot to the new follower node.3. The follower connects to the leader and requests all the data changes that havehappened since the snapshot was taken. This requires that the snapshot is associ‐ated with an exact position in the leader’s replication log. That position has vari‐ous names: for example, PostgreSQL calls it the log sequence number, andMySQL calls it the binlog coordinates.4. When the follower has processed the backlog of data changes since the snapshot,we say it has caught up. It can now continue to process data changes from theleader as they happen
In summary, setting up a new follower in a database system involves creating a consistent snapshot of the leader's data, copying this snapshot to the follower, and then having the follower catch up with the leader by processing all the data changes that occurred since the snapshot was taken. This approach ensures data consistency without the need to lock the entire database, thus maintaining high availability.
materialized view
A materialized view is different from a regular view in that it is a physical copy of the data, while a view is a virtual table that does not store data. When a query is run on a view, it retrieves the data from the original tables, whereas a materialized view stores the results of the query, allowing for faster data retrieval. Materialized views are particularly useful when data needs to be accessed frequently and the underlying data does not change often, as they can improve performance by providing precomputed results.
def find_anagrams(s): if len(s) <= 1: return {s} anagrams = set() for i, char in enumerate(s): # Avoid duplicating work for identical characters if s.index(char) == i: for anagram in find_anagrams(s[:i] + s[i+1:]): anagrams.add(char + anagram) return anagrams # Example usage string = "abc" anagrams = find_anagrams(string) print(anagrams)
**Permutations of a string **
using non-linear color jittering augmentations
image processing techniques that alter the colors of an image in a non-linear manner.
. For example, increasing the contrast of an image can make the darker colors darker and the brighter colors brighter, which can change the perceived illumination color. Similarly, increasing the saturation of an image can make the colors more vivid, which can also change the perceived illumination color. Therefore, it is important to use linear color space when performing the color constancy task
Another line of research, not pursued in thispaper, focusses on the invariance that can be obtained byapplying various photometric transformations, sometimes alsoreferred to as color constancy [19, 20, 21]. Such methods areoften extended to incorporate other forms of invariance, likeinvariance to highlights or shadows, but do not result in outputimages that have any visual similarity to the original inputimage.
Photometric transformations are operations that change the brightness, contrast, color, or intensity of an image, without affecting its geometry
For a robust color-basedsystem, these effects of the light source should be filtered out.
So light can impact color constancy
The arithmetic to convert a duration into hours, minutes and seconds is the same as that in the one-argument constructor of the HmsDuration class. The last line above is identical to that in the asHms implementation of the HmsDuration class. This is not altogether surprising, but as classes get bigger even more code replication is likely. Code replication is not desirable for a variety of reasons: The code is more verbose and repetitive than necessary. If a snippet of code has a bug, it is likely the bug is also replicated at multiple places. Consequently one may fix the bug at some but not all places.
When you design by the interface, code duplication is most likely. To avoid this, we use abstract classes which - Implements the interface - This the base case for other class that used to implement this interface - Can contain methods with/without definition that the child classes can reuse
High-frequenc
high-frequency information refers to the rate of change of intensity values from one pixel to the next. A high-frequency image is one where the intensity values change quickly from one pixel to the next, such as edges in an image.
Transformer for Single Image Super-Resolution
General Comments - Not the first transformer + SISR - Not SOTA - A more efficient approach: the best trade-off between model performance and computation cost.
Critique - Choice of metrics, subjective. Lack of other comparison with other metrics. - Does this compare to the state-of-the-art? Why only compare it to the lightweight ones? How will I know what I am trading off for this?
Inter Pulse Interval (IPI)
How doe IPI vary within each coda click?