झूठ बोल रही है सबसे.....🥺

झूठ बोल रही है सबसे ,सच क्या है तूने बताया नही! 

साथ रहने का किया जो वादा, वो वादा तूने निभाया नही!

 गली मे तो तू आना चाहती थी ,लेकिन मैने तुझे बुलाया नही! 

भूल गई तू मेरी मोहब्बत लेकिन तेरी बेवफाई को मैने भुलाया नही!

Ye Ham Kaha Aagye....🫴🏻

Lahu ke the jo rishtey unhe chhod ke aa gye,

Sukun ankho ke samne tha ham muh mod ke aa gye,

 khajane lut rhe the maa baap ki chaon me,

 chand kaudiyo ke khatir ham ghar chhod ke aa gye.❤️🙌 

Ghar....❤️

Ki bheegi huii ankho ka manjar naa milega... 

Ghar chor ke mat jao kahi ghar na milega.... 

मामूली सा इंसान हूँ....😌

 

मामूली सा इंसान हूँ, 

ज़्यादा मत परखना मुझको, 

हो सके तो इनकार कर देना, 

पर इंतज़ार में मत रखना मुझको। 😍

Bs itna he dukh hai 😢

 

Zara c zindagi h, armaan boht h

humdard koi nhi, insaan boht h

dil ka dard sunaye to sunaye kisko

jo dil ke kareeb h, anjaan boht h

Tere siwa koi jachta kyu nhi❤️

Tere siwa or koi mujhe koi jachta kyu nhi,

tera chehra mere jhan se hatta kyun nhi ,

teri har ek chij se mohbbat hai mujhe 

bs  itni si baat tu samjhta kyu nhi ❤️

Kya lagti h Wo Meri?

Wo mujh ko mera chand lagti hai,

 Durr hai mujhse par mere pas lagti hai, 

Aise to bahut hai mere kareeb par, 

Wo sab se khass lagti hai, 

Usse dost kehdo ya pyaar kehdo mera,

Wo mere harr sawal ka jawab lagti hai

Agar tum gulab hoti☺️❤️

Main Tod leta agar tum gulab hoti

Main jawab ban jata agar tum sawal hoti 

Sab jante Hain ham Nasha nahin karte

Fir bhi Pi leta agar tum sharab hoti☺️❤️

tum baat krogi kya

 Ki tere ghr aa jaunga pr tum milogi kya,,, 

Tumse baatein h krni bahut pr tum baat krogi kya,,, 

Mera intezaar ka imtehaan mat le ae gaalib,,, 

Main to umrabhar yahin hoon, 

magar tum vapas mere paas aaogi kya

तुम मेरे हो बात खतम.. ❤️🙌

इश्क घुला और जात खतम... 

चांद खिला और रात खतम... 

नए दौर की प्रेम कहानी... 

जिस्म मिला जज्बात खतम... 

रिश्ता बिस्ता तुम जानो... 

तुम मेरे हो बात खतम.. ❤️🙌

अब नहीं...❤️

सुधर सुधर के सुधरा हूँ

 मैं फ़िर से बिगड़ जाऊँगा 

तुम पूछोगे हाल मेरा

 मैं इश्क़ में पड़ जाऊँगा

Graph Neural Networks (GNNs)

Excited to share insights from my recent seminar on Graph Neural Networks (GNNs) - Revolutionizing AI! 🤖 Graph Neural Networks in AI: where data meets connectivity. I explored the profound impact of GNNs in transforming AI applications, sparking conversations on their ethical implications alongside live demonstrations. Let's champion a future where advanced AI technologies drive meaningful innovation. 🚀 Report on Graph Neural Networks (GNNs) Introduction to Graph Neural Networks Graph Neural Networks (GNNs) are a class of deep learning methods designed to perform inference on data described by graphs. Unlike traditional neural networks, which operate on structured data like grids or sequences, GNNs are specifically designed to handle graph-structured data, capturing the dependencies and relationships between nodes (entities) and edges (connections). Overview of GNNs Graph Neural Networks aim to leverage the structure of graph data to learn high-level representations. These representations can be used for various tasks, such as node classification, link prediction, and graph classification. GNNs achieve this by iteratively aggregating and transforming node features from their neighbors, propagating information through the graph. Types of GNNs Recurrent Graph Neural Networks (RecGNNs): RecGNNs are among the earliest types of GNNs. They apply recurrent mechanisms to update node states until convergence. These networks use a diffusion mechanism to exchange information between nodes until a stable state is reached. An example is the model proposed by Scarselli et al., which can handle various graph types, including directed, undirected, cyclic, and acyclic graphs. Convolutional Graph Neural Networks (ConvGNNs): ConvGNNs are inspired by Convolutional Neural Networks (CNNs) and apply convolution operations to graph data. The Graph Convolutional Network (GCN) proposed by Kipf and Welling is a prominent example. It generalizes the convolution operation to graph-structured data by aggregating feature information from a node's local neighborhood. Graph Autoencoders (GAEs): GAEs are used for unsupervised learning tasks on graphs. They aim to learn low-dimensional representations of nodes or entire graphs by encoding the input graph into a latent space and then reconstructing it. Variants include Variational Graph Autoencoders (VGAEs), which introduce probabilistic elements into the encoding process. Spatial-Temporal Graph Neural Networks (STGNNs): STGNNs extend GNNs to handle dynamic graph data, where the graph structure and node features evolve over time. These networks are particularly useful for tasks like traffic forecasting, where the data is both spatially and temporally dependent. Key Techniques in GNNs Message Passing: In the message passing framework, nodes iteratively update their representations by aggregating messages from their neighbors. This framework underpins many GNN variants, allowing them to capture complex dependencies in the graph. Pooling Mechanisms: Graph pooling aims to reduce the size of the graph while preserving its essential structure. Techniques like DIFFPOOL introduce a differentiable pooling mechanism that clusters nodes in a hierarchical manner, enabling the construction of deeper GNN models capable of learning hierarchical graph representations. Attention Mechanisms: Attention mechanisms in GNNs allow the model to weigh the importance of different nodes' contributions when aggregating information. This can lead to more effective learning, particularly in heterogeneous graphs where different nodes may have varying degrees of influence. Applications of GNNs GNNs have found applications across various domains due to their ability to model relational data effectively: Social Network Analysis: GNNs are used to predict user attributes, detect communities, and recommend content by analyzing the connections and interactions between users in social networks. Molecular Chemistry: In drug discovery and materials science, GNNs help predict molecular properties by modeling the interactions between atoms within molecules. Traffic and Transportation: STGNNs forecast traffic conditions and optimize transportation networks by considering the temporal and spatial dependencies of traffic data. Recommendation Systems: GNNs enhance recommendation systems by leveraging user-item interaction graphs to provide personalized suggestions. Knowledge Graphs: GNNs are employed to complete and reason over knowledge graphs, facilitating tasks such as question answering and entity recognition. Conclusion Graph Neural Networks represent a powerful and flexible approach to learning from graph-structured data. By extending traditional neural network operations to graphs, GNNs capture the rich relational information inherent in many real-world datasets. The ongoing development of novel GNN architectures and techniques continues to expand their applicability and effectiveness across a growing array of domains. For further details and advanced reading on specific GNN models and their implementations, please refer to the comprehensive resources available in the provided document.

दिन को रात...❤️

 वो अपनी जुल्फे से दिन को रात करती है. 

उसकी आँखे ता उम्र के वादे करती है 

हवाये झूमती है उसे छू कर 

लकिन वो बस मुस्कुराके बरसते करती है

Jariya dhoondta hoon😔

Har roz

 tumse baat karne ka jariya dhoondta hoon,

har raat chaand se teri hi baate karta hoon

pata hai kabhi miloge nahi tum

phir bhi tere aane ka intezaar karta hu

Ek Ajnabi❤️

khawabon ki duniya me

hum khote gye

hosh me they phir bhi madhosh hote gye

 na jane kya jadu tha

uss ajnabi chere me 

khud ko bohot roka phir bhi uske hote gye

मोहब्बत❤️

 

Yeh mohabbat ke haadse aksar dilon ko tor dete hain,

 Tum manzil ki baat karte ho, log raahon mein chhor dete hai

चाँद और तुम💖

अपनी निगाहों से न देख खुदको 

हीरा भी तुझे पत्थर लगेगा

 सब कहते होंगे 

चाँद का टुकड़ा है तू 

मेरी नज़र से देख

चाँद भी तेरा टुकड़ा लगेगा

Dil Ko Chhupaaya Na Karo💓

Dil ko qaid kar ke muskuraya na karo

Roshni ki khatir dil jalaya na karo

 Kyun karte ho dil par itna sitam

Agar yaad nahi karte, tov yaad aaya na karo