The optimization field suffers from the metaphor-based “pseudo-novel” or “fancy” optimizers. Most of these cliché methods mimic animals' searching trends and possess a small contribution to the optimization process itself. Most of these cliché methods suffer from the locally efficient performance, biased verification methods on easy problems, and high similarity between their components' interactions. This study attempts to go beyond the traps of metaphors and introduce a novel metaphor-free population-based optimization method based on the mathematical foundations and ideas of the Runge Kutta (RK) method widely well-known in mathematics. The proposed RUNge Kutta optimizer (RUN) was developed to deal with various types of optimization problems in the future. The RUN utilizes the logic of slope variations computed by the RK method as a promising and logical searching mechanism for global optimization. This search mechanism benefits from two active exploration and exploitation phases for exploring the promising regions in the feature space and constructive movement toward the global best solution. Furthermore, an enhanced solution quality (ESQ) mechanism is employed to avoid the local optimal solutions and increase convergence speed. The RUN algorithm's efficiency was evaluated by comparing with other metaheuristic algorithms in 50 mathematical test functions and four real-world engineering problems. The RUN provided very promising and competitive results, showing superior exploration and exploitation tendencies, fast convergence rate, and local optima avoidance. In optimizing the constrained engineering problems, the metaphor-free RUN demonstrated its suitable performance as well. The authors invite the community for extensive evaluations of this deep-rooted optimizer as a promising tool for real-world optimization

The foundation models have recently shown excellent performance on a variety of downstream tasks in computer vision. However, most existing vision foundation models simply focus on image-level pretraining and adpation, which are limited for dynamic and complex video-level understanding tasks. To fill the gap, we present general video foundation models, InternVideo, by taking advantage of both generative and discriminative self-supervised video learning. Specifically, InternVideo efficiently explores masked video modeling and video-language contrastive learning as the pretraining objectives, and selectively coordinates video representations of these two complementary frameworks in a learnable manner to boost various video applications. Without bells and whistles, InternVideo achieves state-of-the-art performance on 39 video datasets from extensive tasks including video action recognition/detection, video-language alignment, and open-world video applications. Especially, our methods can obtain 91.1% and 77.2% top-1 accuracy on the challenging Kinetics-400 and Something-Something V2 benchmarks, respectively. All of these results effectively show the generality of our InternVideo for video understanding. The code will be released at https://github.com/OpenGVLab/InternVideo.

InternVideo

**Policy Similarity Metric**, or **PSM**, is a similarity metric for measuring behavioral similarity between states in reinforcement learning. It assigns high similarity to states for which the optimal policies in those states as well as in future states are similar. PSM is reward-agnostic, making it more robust for generalization compared to approaches that rely on reward information.

Policy Similarity Metric

Contrastive Behavioral Similarity Embeddings for Generalization in Reinforcement Learning

**PP-OCR** is an OCR system that consists of three parts, text detection, detected boxes rectification and text recognition. The purpose of text detection is to locate the text area in the image. In PP-OCR, Differentiable Binarization (DB) is used as text detector which is based on a simple segmentation network. It integrates feature extraction and sequence modeling. It adopts the Connectionist Temporal Classification (CTC) loss to avoid the inconsistency between prediction and label.

PP-OCR

PP-OCR: A Practical Ultra Lightweight OCR System

A **ShuffleNet Block** is an image model block that utilises a [channel shuffle](https://paperswithcode.com/method/channel-shuffle) operation, along with depthwise convolutions, for an efficient architectural design. It was proposed as part of the [ShuffleNet](https://paperswithcode.com/method/shufflenet) architecture. The starting point is the [Residual Block](https://paperswithcode.com/method/residual-block) unit from [ResNets](https://paperswithcode.com/method/resnet), which is then modified with a pointwise group [convolution](https://paperswithcode.com/method/convolution) and a channel shuffle operation.

ShuffleNet Block

ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices

**node2vec** is a framework for learning graph embeddings for nodes in graphs. Node2vec maximizes a likelihood objective over mappings which preserve neighbourhood distances in higher dimensional spaces. From an algorithm design perspective, node2vec exploits the freedom to define neighbourhoods for nodes and provide an explanation for the effect of the choice of neighborhood on the learned representations. 

For each node, node2vec simulates biased random walks based on an efficient network-aware search strategy and the nodes appearing in the random walk define neighbourhoods. The search strategy accounts for the relative influence nodes exert in a network. It also generalizes prior work alluding to naive search strategies by providing flexibility in exploring neighborhoods.

node2vec

node2vec: Scalable Feature Learning for Networks

**Dot-Product Attention** is an attention mechanism where the alignment score function is calculated as: 

$$f_{att}\left(\textbf{h}_{i}, \textbf{s}\_{j}\right) = h\_{i}^{T}s\_{j}$$

It is equivalent to [multiplicative attention](https://paperswithcode.com/method/multiplicative-attention) (without a trainable weight matrix, assuming this is instead an identity matrix). Here $\textbf{h}$ refers to the hidden states for the encoder, and $\textbf{s}$ is the hidden states for the decoder. The function above is thus a type of alignment score function. 

Within a neural network, once we have the alignment scores, we calculate the final scores/weights using a [softmax](https://paperswithcode.com/method/softmax) function of these alignment scores (ensuring it sums to 1).

Dot-Product Attention

Effective Approaches to Attention-based Neural Machine Translation

FINCH is a parameter-free fast and scalable clustering algorithm. it stands out for its speed and clustering quality.

FINCH Clustering

Efficient Parameter-Free Clustering Using First Neighbor Relations

CDCC-NET is a multi-task network that analyzes the detected counter region and predicts 9 outputs: eight float numbers referring to the corner positions (x0/w, y0/h, ... , x3/w, y3/h) and an array containing two float numbers regarding the probability of the counter being legible/operational or illegible/faulty.

CDCC-NET

Towards Image-based Automatic Meter Reading in Unconstrained Scenarios: A Robust and Efficient Approach

**Video Panoptic Segmentation Network**, or **VPSNet**, is a model for video panoptic segmentation. On top of UPSNet, which is a method for image panoptic segmentation, VPSNet is designed to take an additional frame as the reference to correlate time information at two levels: pixel-level fusion and object-level tracking. To pick up the complementary feature points in the reference frame, a flow-based feature map alignment module is introduced along with an asymmetric attention block that computes similarities between the target and reference features to fuse them into one-frame shape. Additionally, to associate object instances across time, 
 an object track head is added which learns the correspondence between the instances in the target and reference frames based
on their RoI feature similarity.

VPSNet

Video Panoptic Segmentation

**SEED** (Scalable, Efficient, Deep-RL) is a scalable reinforcement learning agent. It utilizes an architecture that features centralized inference and an optimized communication layer. SEED adopts two state of the art distributed algorithms, [IMPALA](https://paperswithcode.com/method/impala)/[V-trace](https://paperswithcode.com/method/v-trace) (policy gradients) and R2D2 ([Q-learning](https://paperswithcode.com/method/q-learning)).

SEED RL

SEED RL: Scalable and Efficient Deep-RL with Accelerated Central Inference

**OA-Loss** reduces the domain gap between the original and augmented images. OA-Loss enables the model to learn semantic relations of foreground and background instances from multi-domain.

OA-Loss

Object-Aware Domain Generalization for Object Detection

A **Bottleneck Transformer Block** is a block used in [Bottleneck Transformers](https://www.paperswithcode.com/method/bottleneck-transformer) that replaces the spatial 3 × 3 [convolution](https://paperswithcode.com/method/convolution) layer in a [Residual Block](https://paperswithcode.com/method/residual-block) with Multi-Head Self-Attention (MHSA).

Bottleneck Transformer Block

Bottleneck Transformers for Visual Recognition

**MPNet** is a pre-training method for language models that combines masked language modeling (MLM) and permuted language modeling (PLM) in one view. It takes the dependency among the predicted tokens into consideration through permuted language modeling and thus avoids the issue of [BERT](https://paperswithcode.com/method/bert). On the other hand, it takes position information of all tokens as input to make the model see the position information of all the tokens and thus alleviates the position discrepancy of [XLNet](https://paperswithcode.com/method/xlnet).

The training objective of MPNet is:

$$ \mathbb{E}\_{z\in{\mathcal{Z}\_{n}}} \sum^{n}\_{t=c+1}\log{P}\left(x\_{z\_{t}}\mid{x\_{z\_{<t}}}, M\_{z\_{{>}{c}}}; \theta\right) $$

As can be seen, MPNet conditions on ${x\_{z\_{<t}}}$ (the tokens preceding the current predicted token $x\_{z\_{t}}$) rather than only the non-predicted tokens ${x\_{z\_{<=c}}}$ in MLM; comparing with PLM, MPNet takes more information (i.e., the mask symbol $[M]$ in position $z\_{>c}$) as inputs. Although the objective seems simple, it is challenging to implement the model efficiently. For details, see the paper.

MPNet

MPNet: Masked and Permuted Pre-training for Language Understanding

We present a novel classifier network called STEP, to classify perceived human emotion from gaits, based on a Spatial Temporal Graph Convolutional Network (ST-[GCN](https://paperswithcode.com/method/gcn)) architecture. Given an RGB video of an individual walking, our formulation implicitly exploits the gait features to classify the perceived emotion of the human into one of four emotions: happy, sad, angry, or neutral. We train STEP on annotated real-world gait videos, augmented with annotated synthetic gaits generated using a novel generative network called STEP-Gen, built on an ST-GCN based Conditional Variational Autoencoder (CVAE). We incorporate a novel push-pull regularization loss in the CVAE formulation of STEP-Gen to generate realistic gaits and improve the classification accuracy of STEP.
We also release a novel dataset (E-Gait), which consists of 4,227 human gaits annotated with perceived emotions along with thousands of synthetic gaits. In practice, STEP can learn the affective features and exhibits classification accuracy of 88\% on E-Gait, which is 14--30\% more accurate over prior methods.

STEP: Spatial Temporal Graph Convolutional Networks for Emotion Perception from Gaits

A Gsop block has a squeeze module and an excitation module, and uses a second-order pooling to model high-order statistics while gathering global information.
In the squeeze module, a GSoP block firstly reduces the number of channels from $c$ to $c'$ ($c' < c$) using a $1 \times 1$ convolution,  then  computes a $c' \times c'$ covariance matrix for the different channels to obtain their correlation.  Next, row-wise normalization is performed on the covariance matrix.  Each $(i, j)$ in the normalized covariance matrix explicitly relates channel $i$ to channel $j$. 

In the excitation module, a GSoP block performs row-wise convolution to  maintain structural information and output a vector. Then a fully-connected layer and a sigmoid function are applied  to get a $c$-dimensional attention vector. Finally, it multiplies the input features by the attention vector, as in an SE block. A GSoP block can be formulated as:
\begin{align}
    s = F_\text{gsop}(X, \theta) & = \sigma (W \text{RC}(\text{Cov}(\text{Conv}(X))))
\end{align}
\begin{align}
    Y & = s  X
\end{align}
Here, $\text{Conv}(\cdot)$ reduces the number of channels,
$\text{Cov}(\cdot)$ computes the covariance matrix and
$\text{RC}(\cdot)$ means row-wise convolution.

GSoP-Net

Global Second-order Pooling Convolutional Networks

**Receptive Field Block (RFB)** is a module for strengthening the deep features learned from lightweight CNN models so that they can contribute to fast and accurate detectors. Specifically, RFB makes use of multi-branch pooling with varying kernels corresponding to RFs of different sizes, applies [dilated convolution](https://paperswithcode.com/method/dilated-convolution) layers to control their eccentricities, and reshapes them to generate
final representation.

Receptive Field Block Net for Accurate and Fast Object Detection

A **Wide Residual Block** is a type of [residual block](https://paperswithcode.com/method/residual-block) that utilises two conv 3x3 layers (with [dropout](https://paperswithcode.com/method/dropout)). This is wider than other variants of residual blocks (for instance [bottleneck residual blocks](https://paperswithcode.com/method/bottleneck-residual-block)). It was proposed as part of the [WideResNet](https://paperswithcode.com/method/wideresnet) CNN architecture.

AI Glossary

Rung Kutta optimization

InternVideo: General Video Foundation Models via Generative and Discriminative Learning

Policy Similarity Metric

PP-OCR

ShuffleNet Block

node2vec

Dot-Product Attention

First Integer Neighbor Clustering Hierarchy (FINCH))

CDCC-NET

Video Panoptic Segmentation Network

SEED RL

Object-Aware Loss

Bottleneck Transformer Block

MPNet

Gait Emotion Recognition

Global second-order pooling convolutional networks

Receptive Field Block

Wide Residual Block